Contents
Behavioural Ethics & Decision-Making
Section 1
Core Idea
Omission bias is the systematic preference for harm caused by not acting over harm caused by acting, even when the outcomes are identical or the inaction produces worse results. It was formalised by Ilana Ritov and Jonathan Baron in a 1990 series of vaccination studies: subjects were asked whether they would vaccinate a hypothetical child against a disease when the vaccine itself carried a small risk of causing exactly the harm it prevented. Even when the vaccine strictly reduced the child's overall risk of harm, a substantial minority refused, on the grounds that a death caused by the vaccine felt worse than a mathematically identical death caused by the disease. The action was aversive in a way the omission was not.
The bias survives across domains because its psychological grammar is durable. Actions are attributed to the actor; omissions get diffused across circumstance, fate, or the general run of the world. If you did something and it went badly, you are the cause. If you did nothing and the same thing went badly, the event has causes but you are not among them, at least in the folk-morality most of us actually operate under. This is not a defensible ethical position on inspection — a doctor who withholds a plainly life-saving treatment is not innocent of the death — but the psychological asymmetry is real, and it shapes decisions before ethical reasoning gets a chance to catch up.
The bias also compounds with loss aversion and status-quo bias. Losses feel roughly twice as large as equivalent gains (Kahneman and Tversky's prospect theory), which means a bad outcome from an action registers as a much larger loss than the same bad outcome from inaction, which barely registers as a loss at all. And the status quo is the reference point from which "action" and "omission" are defined in the first place. The two together mean that the default of doing nothing carries a hidden discount that most decision-makers never explicitly price.
For operators and leaders, this is one of the most expensive biases in ordinary business life, because it is invisible in the incident reports. Nobody writes a post-mortem about the deal you didn't do, the launch you delayed, or the hire you passed on. The costs of omission are counterfactual, and counterfactuals don't fire people.
Section 2
How to See It
Medical Decisions
You're seeing it when patients or families accept a higher-mortality "let nature take its course" path over a lower-mortality intervention with a rare side-effect. The interventional death would feel like the doctor's fault; the natural-history death feels like the disease's. The math is the same; the moral accounting isn't.
Corporate Risk
You're seeing it when a leadership team declines to make a decision they can obviously see is correct, choosing instead to "wait and see." The wait costs more than the action would — but the wait doesn't show up on anyone's ledger, and the action would have. The default is not neutrality; it is a decision to accept the costs of inaction.
Public Policy
You're seeing it when regulators approve a low-benefit, low-risk intervention faster than a high-benefit, higher-risk one, because the accountability for approving harm is asymmetric with the accountability for delaying benefit. The FDA's historical asymmetry on Type I versus Type II errors is a canonical example.
Section 3
How to Use It
The countermove is to explicitly price omissions. Treat the choice to do nothing as an active decision with a full cost column, not as the natural default from which action must be justified. This is the discipline Bezos wrote into Amazon's culture with his framing of Type 1 (irreversible) versus Type 2 (reversible) decisions: for reversible decisions, the cost of not deciding usually dominates the cost of a bad decision, because the bad decision can be corrected while the delay cannot be recovered.
Two concrete techniques:
Pre-mortem the omission, not just the action. Before choosing inaction, force yourself to write the post-mortem of the world in which you did not act and something bad happened as a result. If that post-mortem is more painful than the one where you acted and something bad happened, the omission is not actually the safer path — it is just the path with less visible ownership.
Name the counterfactual you are accepting. "By not launching this product, we are accepting an expected value of losing X% of the market to competitor Y, over Z months, with the following knock-on effects." When the counterfactual is written down explicitly, the omission stops feeling free.
Decision filter
"If both the action and the inaction produced the identical bad outcome, would I feel equally responsible? If not, which side of that asymmetry is doing the deciding — the ethics or the accounting?"
As a founder
The single most expensive habit in early-stage companies is deferred decisions — hires you don't make, PMs you don't fire, features you don't cut, pricing you don't raise. None of these deferrals show up on a report card. All of them compound. The specific counter-discipline is to time-box every non-decision: "we will decide this by Friday, and 'no decision' is not on the menu." The forcing function overrides the omission asymmetry.
As a manager
When you find yourself repeatedly presented with a subordinate's underperformance and repeatedly choosing not to act, you are not being kind. You are trading a small acute cost (the difficult conversation) for a large chronic cost (a team that watches you tolerate what you have publicly said matters). Omission is not the neutral option in management; it is the option with the highest half-life.
Section 4
Common Misapplications
Three ways the model gets misused:
-
Confusing action bias with wisdom. The correction to omission bias is not "always act." Some decisions really should be delayed pending information, and some real virtues — patience, epistemic humility, respect for reversibility — do favour inaction. The right frame is symmetric accounting, not a thumb on the action scale. A well-priced omission is exactly as valid as a well-priced action.
-
Ignoring reversibility asymmetry. Actions and omissions are not always symmetric in their consequences. An irreversible action (a firing, a bet-the-company acquisition, a war) is genuinely more expensive than the corresponding omission, because it forecloses future options. When the action is truly irreversible, some extra weight on the omission side is not bias but prudence. The bias arises when this reasoning is applied to reversible decisions, which is where most operational choices actually sit.
-
Using "we didn't do it" as a moral shield. In institutional life, the language of omission is often deployed as legal or reputational protection — "the company did not cause this outcome," "we merely failed to prevent it." That framing may hold in court; it rarely holds ethically. If you had the information, the capacity, and a reasonable duty to act, the omission is a decision, and describing it in the passive voice does not exempt you from ownership of it. This is the cowardice-with-better-PR version of the bias, and it is common enough in corporate and political life to deserve a name of its own.
Section 5
Founders & Leaders
Jeff BezosFounder, Amazon
Bezos's "Type 1 vs Type 2 decisions" framing is a direct institutional counter to omission bias. His argument, made repeatedly in shareholder letters, is that for reversible (Type 2) decisions, the cost of inaction is almost always greater than the cost of an imperfect action, because delay itself has an opportunity cost that compounds. The framework does not encourage recklessness; it explicitly reserves careful deliberation for the irreversible cases, which is exactly the right sorting of when omission bias is a risk versus a virtue.
Reed HastingsFounder, Netflix
Hastings' willingness to publicly cannibalise Netflix's DVD business with streaming — and to accept the near-term stock-price penalty for doing so — is a decisive rejection of omission bias at CEO scale. The comfortable choice was to defer the transition and let the DVD business run down organically. Every year of delay would have been a year of quiet omission. The 2011 pivot booked the cost up front.
Section 6
Company Examples
Netflix
The DVD-to-streaming transition is the paradigm corporate example of overriding omission bias. Every internal signal — near-term revenue, subscriber inertia, executive relationships with retail partners — argued for delay. The counterfactual cost of continuing to do nothing (extinction at the hands of a faster-moving competitor) was diffuse and years away. Hastings priced it explicitly and acted. The subsequent decade of growth is a monument to the value of pricing omissions correctly.
T
Tesla
Tesla's early willingness to release imperfect vehicles and iterate is a countercultural rejection of the automotive industry's omission bias, which historically preferred to defer risky launches indefinitely rather than ship a product with a known defect. The industry norm — treating the omission (not launching) as free while the action (launching an imperfect car) carries full accountability — is exactly the asymmetry Ritov and Baron described. Tesla's ability to price both sides is a large part of the company's speed advantage, and also of its safety-related controversies. The trade is real, and it belongs on both sides of the ledger.
Section 7
Connected Models
Pairs-with
Status Quo Bias
Status quo bias defines the reference point from which action and inaction are labelled. Omission bias then weights them asymmetrically.
Amplified by
Loss Aversion
Losses from action loom larger than losses from inaction, even when the outcomes are identical.
Counterweighted by
Action Bias
In some domains — notably medicine and management — the mirror-image bias appears, where actors intervene when doing nothing would have been better. Both biases exist; neither is a general prescription.
Interacts with
Reversibility
For irreversible decisions, extra weight on omission is prudence, not bias. The model is most costly for reversible decisions treated as irreversible.
Section 8
One Key Quote
— Ilana Ritov & Jonathan Baron, Journal of Behavioral Decision Making, 1990"Subjects tended to judge acts that resulted in harmful outcomes as worse than omissions that resulted in identical outcomes — an asymmetry that persists even when the omission is objectively more harmful."
Section 11
Summary & Further Reading
Omission bias is the systematic under-pricing of the costs of inaction. It is powered by the psychological asymmetry between actions attributed to actors and outcomes attributed to fate. Its everyday form in business is the deferred decision, the un-fired underperformer, the un-launched product, and the un-had conversation. The countermove is symmetric accounting: price omissions with the same rigour as actions, distinguish reversible from irreversible decisions, and treat "no decision" as a decision with a full cost column. Good judgement is not a preference for action; it is the discipline of pricing both sides of the ledger honestly.
01
Paper
The 1990 experimental paper that formalised omission bias. Concise, replicated widely, and still the primary reference for the effect.
02
Book
Kahneman's synthesis of prospect theory places omission bias in its broader psychological context — loss aversion, status-quo bias, and the machinery of intuitive judgement.
03
Article
The 2015 and 2016 letters contain the clearest public articulation of the Type 1 / Type 2 decision framework — the working operator's antidote to omission bias.
Why this matters next
mental modelsInertia
Omission Bias applied the Inertia mental model
mental modelsScale
Omission Bias applied the Scale mental model
mental modelsHalf-life
Omission Bias applied the Half-life mental model
mental modelsCounterfactuals
Omission Bias applied the Counterfactuals mental model
mental modelsCost
Omission Bias applied the Cost mental model
mental modelsForcing Function
Omission Bias applied the Forcing Function mental model
Frequently asked questions
What is Omission Bias?
Omission Bias is a mental model used for better thinking and decision-making.
How do you apply Omission Bias?
To apply Omission Bias, identify situations where this framework is relevant, then use it as a lens to evaluate your options and decisions. The model is most useful when combined with other complementary mental models.
What category does Omission Bias fall under?
Omission Bias falls under the Psychology & Behavior category of mental models. Other models in this category can be found on the Psychology & Behavior hub page.
Why is Omission Bias important?
Omission Bias is important because it provides a structured way to think about problems that would otherwise be approached with intuition alone. Understanding this model helps you avoid common reasoning errors and make better decisions.
Continue exploring
AHANBEBECDCB
Mental model
Availability Heuristic
We judge the likelihood of events based on how easily examples come to mind, not
Mental model
Anchoring
The first piece of information encountered disproportionately influences all sub
Mental model
Bandwagon Effect
People adopt beliefs, behaviours, and trends simply because others have — popula
Mental model
Bystander Effect
The more people present during an emergency, the less likely any individual is t
Mental model
Cognitive Dissonance
When beliefs and actions contradict, the resulting psychological discomfort driv
Mental model
Confirmation Bias
The tendency to seek, interpret, and remember information that confirms existing
More like this, in your inbox
I send a newsletter every week — free, no spam, unsubscribe anytime.