I'm having trouble reconciling these two biases. They seem to be the exact opposite of each other. Yet humans are allegedly pre-disposed to both.

How can they be reconciled?

Omission Bias:

The omission bias is an alleged type of cognitive bias. It is the tendency to judge harmful actions as worse, or less moral than equally harmful omissions (inactions) because actions are more obvious than inactions.

Action Bias (academic source: Bar-Eli et al., 2007)

Simply put this means that just about everyone, when faced with ambiguous situations, especially those circumstances associated with risk, gets the feeling that they need to take some action regardless of whether this is a good idea or not.

  • 2
    $\begingroup$ Please define the biases and their use in you context. $\endgroup$
    – BB King
    Dec 27 '15 at 19:22
  • $\begingroup$ I don't think there is anything to reconcile here at all. These biases don't really relate to economics, but to morality, so I am not sure where your confusion comes from. $\endgroup$
    – rocinante
    Dec 28 '15 at 5:52
  • $\begingroup$ @rocinante Please tell that to Bar-Eli et al. Their paper was in the Journal of Economic Psychology. Also, this is a question about Behavioral Economics. Google it if it's unfamiliar to you. $\endgroup$ Dec 28 '15 at 7:20
  • 2
    $\begingroup$ @user1883050 While I tend to agree with your last comment, it's also a tad rude. In my mind, these observed biases exist in the realm of psychology until a behavioral economist comes along, loosens one of the homo economicus assumptions, thereby "rationalizing'' the behavior (the quotation marks being critical, as weakening those assumptions is a departure from rationality). So while I think it's a fine question for this forum, don't be surprised by a little backlash. Many economists, myself included, think most behavioral economics is deeply flawed. $\endgroup$
    – Shane
    Dec 28 '15 at 15:36
  • $\begingroup$ ... To give you a flavor of the issues with behavioral economics, many papers weaken assumptions or come up with some poorly motivated model that can explain a particular observed behavior. But, if the model is totally counter-intuitive and explains only one phenomenon, then, in constructing the model, you've basically just assumed the result you were trying to show. Better behavioral papers have intuitive/plausible models and explain several observed behavioral phenomena. $\endgroup$
    – Shane
    Dec 28 '15 at 15:42

According to the reference cited, omission bias arises when the norm for a particular situation is inaction whereas action bias arises when the norm is action. Make sense?

An example of Action Bias in Finance would be the observation that managers often make unprofitable investments instead of the alternative (i.e. return money to investors either as dividends or share buybacks) [1] because for these managers the norm is to find and invest in new projects.

As far as Omission Bias goes, one of the most famous examples arises in this thought experiment.

[1] An example of this is described by Michael C. Jensen in 'The Agency Costs of Free Cash Flow: Corporate Finance and Takeovers' [American Economic Review, Vol. 76, No. 2 (May, 1986)].


This is a great question, you're not the only one confused. We just completed a working paper on this very topic: "What is action, what is inaction? A review of action-inaction biases and recommendations for term use and typology" and we cover these two biases and many other action-inaction biases.

See page 5:

Building on the action-effect is the omission bias. The omission bias has been defined in many different ways, and is sometimes regarded to be the same as the action-effect. We view the two biases as different, with the omission bias extending the action-effect to focus on the choices made between action and inaction or about moral attributions to action and inaction, rather than the experience of negative emotions or regret as a consequence of action-inaction. In its broadest definition, the omission bias is the phenomenon that people prefer omissions to commission when there is the possibility of a negative outcome (Spranca, Minsk, & Baron, 1991; Ritov & Baron, 1990). Using this definition, if we consider the classic action-effect scenario, the two investors facing a risky situation and presented with the options of taking action and switching versus not taking action and sticking to the initial investment would prefer inaction to action. Linking with the action-effect, the bias could partly be explained as a strategy to minimize anticipated-regret (Anderson, 2003; Wong & Kwong, 2007; Zhang & Fishbach, 2005), possibly further heightened by related biases such as loss aversion (Tversky & Kahneman, 1991) and risk aversion (Kahneman & Tversky, 1979; Tversky & Kahneman, 1992).

Briefly, the "action effect" is the tendency to regret actions more than inaction (negative outcomes) leading to a preference for omissions. In Bar-Eli's paper the action-effect is reversed to an inaction-effect (Zeelenberg etal. 2002) due to normality, the expectations for goal keepers to act. Therefore, the reconciliation is by using normality from norm theory (Kahneman & Miller, 1986).

Hope that helps, we'd appreciate feedback if things aren't clear.


Here are two examples that illustrate the potential difference to me:

Example 1: You are a general and 200 of your troops are being held captive by an enemy. All friendly soldiers are equally valued by the general and all deaths below are friendly deaths. The general is indifferent about enemy deaths. He must pick one of the two options:

  1. Attack the enemy, resulting in 200 deaths with certainty.
  2. Do nothing, in which case the enemy will kill the 200 captured soldiers.

Omission bias might lead the general to prefer option 2. The fuzzy human reasoning might be that in choosing the first option, the general somehow causes 200 deaths, while in choosing the second option, it is the enemy that causes the deaths, so the general may prefer the second option even though he should be indifferent between them. This example can not be applied for Action Bias because there is no ambiguity, uncertainty or risk in the example.

Example 2: There is an urn with 1 red ball and 1 blue ball inside. A ball is drawn randomly from the urn, and, if it matches your color, you win $100. Your options are as follows:

  1. Choose red as your color
  2. Choose blue as your color
  3. Do nothing and be randomly assigned a color

Note that Omission bias is inapplicable here as there are no harmful actions, and there are really no morality issues at play. On the other hand, Action bias might imply that agents are more likely to pick the first or second options here than they are to choose the third option, even though homo economicus would be indifferent across the three.

To conclude, the key difference, as I see it, is that Action bias is really only relevant in situations with risk or ambiguity. I think technically speaking there is no ambiguity here, since the rules of the game and the balls in the urn are known to the player, but you could easily transform it into a game with ambiguity by saying the player doesn't know the contents of the urn and instead has some belief. Omission bias, on the other hand, is more related to morality. If a bad thing is going to happen, we'd rather it happen because of something that we didn't do than something that we did.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.