Framing Biases in Plea Bargaining Decisions: Fairness Under the Law

Abstract

The traditional rational choice model assumes that people seek to maximize their own expected utility, without accounting for fairness, justice, or concern for the greater good. Reyna et al. (2025) provide a compelling empirical demonstration of how this model breaks down in the context of plea bargaining. Their innovative findings open the door to a wide range of new questions about how framing influences legal decision-making. Although the study focused on general attitudes toward the justice system, future research could build on this foundation by examining how the perceived fairness of a specific plea offer affects decision-making. Whether an offer feels fair is likely to play a central role in its acceptance or rejection. Across a range of contexts, fairness perceptions often shape behavior in ways that depart from the predictions of rational choice theory.

Rational decision-making often breaks down when an offer feels unfair, even if accepting it would serve everyone's best interest. Richard Thaler (2015) offers a particularly memorable example in his book Misbehaving: The Making of Behavioral Economics, illustrating how fairness concerns can interfere with efficient bargaining. The Coase theorem posits that when legal rights are clearly defined and parties can bargain without significant costs, they will negotiate to reach efficient outcomes based on who values the result most. Accordingly, no matter who “owns” something at the start, the final outcome will reflect who cares most about a specific result. Thaler had a tree in his backyard that stood close to the property line and regularly dropped leaves, making cleanup a hassle in the late fall. Whereas Thaler did not care much about the tree one way or another, his neighbor hated it and asked for it to be removed. Applying Coasean logic, Thaler proposed a seemingly efficient solution: Since the neighbor disliked the tree more than Thaler valued keeping it, the neighbor could arrange to have it removed at their own expense. But instead of accepting what a rational actor might see as a fair deal, the neighbor was offended, slammed the door, and never brought up the tree again. Despite being economically rational, the offer struck the neighbor as unfair. After all, it was Thaler's tree, so they felt he should be the one to bear the cost.

A more systematic demonstration of fairness overriding rational self-interest is found in the ultimatum game. In this two-player game, one person (the proposer) offers a split of a sum of money, and the other (the responder) decides whether to accept or reject the offer. If accepted, the money is divided as proposed. If rejected, both players receive nothing. Rational choice theory predicts that responders should accept any amount greater than zero, and proposers, anticipating this, should offer the smallest possible amount. However, this is not how people typically behave. Responders frequently reject low offers they perceive as unfair, even though rejecting means walking away with nothing. For example, if the proposer is given $100, rational models would suggest offering $1 and keeping $99. In practice, however, offers below 20% are often rejected, and proposers tend to offer far more, typically between 40% and 50% (Güth and Kocher 2014).

These examples show that people often resist offers that feel unfair, even at a personal cost. When an offer is perceived as unjust, individuals may reject it in order to preserve a sense of moral or relational equity, even when doing so is irrational by conventional standards. This dynamic has clear relevance for plea bargaining. A defendant may reject a plea not just because of the sentence length or probability of conviction but because the offer itself feels wrong. Moreover, the way the offer is framed—especially as a loss—may strongly influence how unfair it feels.

Consider the following examples that illustrate how framing influences whether an action is perceived as unfair. In a study by Kahneman et al. (1991), participants were asked to judge the fairness of different company decisions. In one scenario, “A company is making a small profit. It is located in a community experiencing a recession with substantial unemployment but no inflation. The company decides to decrease wages and salaries 7 percent this year” (204). In this case, 63% of respondents judged the action to be unfair. In a different scenario, “A company is making a small profit. It is located in a community experiencing a recession with substantial unemployment and inflation of 12 percent. The company decides to increase salaries only 5 percent this year” (204). Here, only 22% judged the action to be unfair. Although both cases result in a comparable real wage decrease, the second is framed as a gain, which elicited significantly lower perceptions of unfairness.

A similar pattern appeared in a pricing example involving car sales. In this case, Kahneman et al. (1991) presented participants with two different scenarios, both beginning with a shortage for a popular car model that would delay delivery by 2 months. In one version, participants read that “a dealer has been selling these cars at list price” and now prices them at $200 above list price; 71% judged this increase as unfair. In the other version, participants read that “a dealer has been selling these cars at a discount of $200 below list price” and now charges list price; only 42% judged this change as unfair (203). Kahneman et al. persuasively argue that “imposing a surcharge (which is likely to be judged a loss) is considered more unfair than eliminating a discount (a reduction of a gain)” (204). The implication is that loss framing can heighten perceptions of unfairness relative to gain framing, even when the options are equivalent.

This insight is directly relevant to the plea offer framing examined in Reyna et al. (2025). For example, one version of the plea offer was framed as “gaining 10 years of freedom,” whereas the other described it as “losing 30 years of freedom.” If the loss-framed version felt more unfair to participants, this could help explain the higher rejection rates observed in that condition. People tend to reject offers that feel unfair, and losses are often perceived as more unfair than comparable gains.

Another important factor likely to influence how fair a plea offer is perceived to be is whether the defendant is actually innocent or guilty. Reyna et al. (2025) deliberately avoided telling participants whether they were to assume that they were innocent or guilty of the crime. This design choice may have been informed by findings from Garnier-Dykstra and T. Wilson (2021), who observed a puzzling reversal in framing effects depending on guilt status. In that study, the standard framing effect (greater plea rejection under a loss frame) appeared when participants were told they were innocent, consistent with Reyna et al.'s results. However, when participants were told they were guilty, the effect reversed: loss-framed offers were more likely to be accepted than gain-framed ones.

Although leaving the defendant's guilt ambiguous could allow for a focused examination of framing effects independent of culpability, defendants in real-world settings almost certainly know whether they committed the crime. That knowledge may strongly influence how fair a plea offer feels. For someone who knows they are innocent, the prospect of jail time is likely to feel fundamentally unfair. Indeed, Tor et al. (2010) found that innocent participants were less likely to accept a plea offer than guilty participants, likely due in part to fairness concerns.

Reyna et al. (2025) offer a methodologically rigorous and conceptually insightful demonstration of how gain and loss framing can shape plea bargain decisions. Their work highlights important departures from traditional rational choice models, particularly in legal contexts where perceptions of fairness and justice often guide behavior. By showing that framing influences defendants' choices, their study provides a valuable foundation for understanding the psychological processes that underlie plea decision-making.

Their findings also invite further exploration into the role of fairness perceptions in this context. Although the study focused on broader attitudes toward the justice system, future research inspired by their work could examine how the perceived fairness of a specific plea offer affects whether it is accepted or rejected. This question is especially relevant in light of existing research showing that people often decline offers they view as unfair, even when accepting would serve their interests.

As behavioral research continues to shape our understanding of legal decision-making, fairness and framing are likely to remain central themes. Investigating how these factors interact, particularly in relation to defendants' knowledge of their own guilt or innocence, may help clarify when and why plea offers are rejected. Building on the insights of Reyna et al. (2025), future work can advance more psychologically informed and ethically grounded models of decision-making in the justice system. Their work lays a strong foundation for future studies examining how fairness perceptions and framing shape decisions in real-world legal settings.