Cognitive Biases Stop Us From Donating Effectively

Distance bias describes the cognitive bias that causes individuals to place greater importance on things that are closer to them, both physically and temporally, even when further things might be equally or more important.⁷ It is why an earthquake in California might elicit a more emotional response from Canadians than an equally disastrous earthquake in Chile. In the historical literature, Peter Singer alluded to distance bias in his famous essay titled Famine, Affluence, and Morality.⁸ Singer poignantly argued that “it makes no moral difference whether the person I can help is a neighbor’s child ten yards away from me or a Bengali whose name I shall never know, ten thousand miles away.” 

If donation recipients are chosen based on proximity instead of necessity, resources may not reach those who need them the most. Global charitable giving is predominantly fuelled by rich countries,⁹ hence why a problematic misallocation of resources can occur when distance bias influences where donations go. If donors in rich countries only donate funds to those who live in their local communities or cities, then deserving recipients from developing countries who would stand to benefit the most will not receive the help they need. The tax-deductible nature of donations magnifies this further; compared to low-income individuals, ultra-wealthy donors (who have high marginal income tax rates) have a comparatively high incentive to donate due to their lower marginal cost of giving.¹⁰

There is empirical evidence that donors generally prefer those nearby to them, as only 31% of donors worldwide choose to donate to charitable organizations located outside of their respective countries of residence, according to survey data.¹¹ Furthermore, experimental evidence shows that as the psychological distance between prospective donors and recipients increases, donors are less willing to provide help, but only when the recipient’s identity remains unknown.¹²

These biases occur because altruists are Humans, not Econs. As described by Cass Sunstein and Richard Thaler in their book Nudge, an Econ is a theoretical type of rational economic decision-maker who makes unbiased forecasts and optimizes given their choices.¹³ By contrast, Humans suffer from innate cognitive biases that limit rationality,¹⁴ especially when faced with ambiguous or probabilistic choices.¹⁵ These biases sometimes cause us to overeat, fail to save, and indulge in vices such as cigarettes or alcohol — activities that an Econ would tend to avoid given the costs involved. While an altruistic Econ would calculate which individuals can be helped the most with each dollar and donate accordingly, a Human might instead choose to donate to charities that are more emotionally striking or familiar to them, even if those charities are not particularly effective or evidence-based.

The second cognitive bias that affects charitable decisions is the identifiable victim effect. This bias describes an individuals’ inclination to be more charitable to a specific, identifiable victim compared to a larger and more ambiguous group with an equal or greater need for help.¹⁶ It is why a television advertisement that shows John, a 5-year old boy with a rare disease, might be significantly more effective at soliciting donations than a similar advertisement that mentions the millions of children who die annually due to insufficient access to clean drinking water.¹⁷ The identifiable victim effect is candidly exemplified by a quote that is commonly attributed to Joseph Stalin: “A single death is a tragedy. A million deaths is a statistic.”

There is empirical evidence that confirms the influence of the identifiable victim effect on donation behavior as seen in Small, Loewenstein, and Slovic’s study from 2007.¹⁸ The study’s participants were each given $5 for completing a short survey, as well as one of three charity request letters. Each of these letters either contained the story of Rokia, a starving girl from Mali; a statistic on how many children are dying from starvation in Mali; or the description of Rokia in addition to the statistical information. Of the three conditions, the letter that featured Rokia’s story alone received the highest average donation.

Most charities are aware of the identifiable victim effect and often feature prominent stories of needy individuals in advertisements. Problems may arise when charities do not receive donations based on their potential to improve lives — the principal concern for Econs when they decide where to donate — but rather by how good they are at leveraging emotions and crafting stories that Humans find captivating. Since charities are often incentivized to raise donations by winning over donors’ hearts, effective charities that do not leverage the identifiable victim effect may do so to the benefit of ineffective ones that do.

Biases and heuristics are hardwired into our brains,¹⁹ hence why making individuals aware of their cognitive biases can sometimes be an ineffective strategy for improving decision making.²⁰ However, charitable organizations can play a prominent role in correcting these biases through the use of properly designed choice architecture, a term that describes the context in which choices are presented to decision-makers.

For the identifiable victim effect, I argue that effective charities should not necessarily try to correct it, but instead leverage it. Researchers have found evidence that an essential driver of the identifiable victim effect is that a higher proportion of those identified can be saved compared to statistical groups.¹⁶ The researchers noted that “When victims are identified it is clear exactly how many people will die, but when victims are statistical it is always possible that more or fewer will die […] Subjects felt avoiding certain fatalities was more important than avoiding uncertain fatalities.”

This ambiguity can potentially be ameliorated by communicating expected values, thereby showing donors that they can indeed make a tangible difference. If, for example, a donation has a 50% chance of saving ten lives, it may be better to communicate that it would save five lives instead of framing it as having the potential — but not the certainty — of saving ten. An even more effective solution would be to harness loss aversion, such that the donation is framed as preventing five deaths versus saving five lives.²¹ Ultimately, these are empirical questions that could benefit from further research that uses randomized controlled trials to test which communications are the most effective under specific circumstances.

A possible way of correcting distance bias could be by reducing the psychological distance between donors and recipients when soliciting donations. Researchers found that framing climate change communications to reduce psychological distance increased mitigation intentions significantly.²² Showing videos of the hypothetical impacts of climate change on the local surroundings of the participants in a treatment group increased willingness to take action to a greater degree than the control group, who instead watched the impacts of climate change on other countries. Similarities between climate change and altruistic decisions — namely, uncertainty regarding the impacts of individual actions, the need to sacrifice in the present, and geographical distance between actors — could serve as a useful analog for identifying strategies to improve the effectiveness of altruistic decisions. 

Based on the aforementioned literature, effective charities could potentially combat distance bias by communicating the evidence-based impacts of their work through referential terms that donors can easily interpret. In New York, saying “Every 10 days, enough people die from malaria to fill Madison Square Garden” might be more salient to donors than “Every 45 seconds, a child dies somewhere in the world from malaria”, thereby reducing distance bias.²³ Another possibility could be to attach an identity to the statistics: “Every 45 seconds, a child similar to Rokia dies in Mali.” Improved decision making can also occur from presenting impacts in terms that donors actually care about: the number of lives saved, the number of people cured of an ailment, or the increase in recipients’ income from an intervention, for example.

Distance bias and the identifiable victim effect can interfere with our ability to make rational altruistic decisions, necessitating effective remediation from charities in the form of well-designed choice architecture. Effective charities should aim to create better solutions that take into account common biases, ultimately resulting in a better alignment of an individual’s desire to help a cause with the behavior necessary to do so. To inform these solutions (and to ensure greater external validity), empirical research that goes beyond experimental environments is needed. Charitable decisions — and the recipients from around the world who stand to gain the most from them — deserve an evidence-based approach that can translate to real-world impact. Behavioral science insights can create transformative social good by better understanding what fuels these decisions for individuals, organizations, and governments alike.