Zero-Risk Bias

What is zero-risk bias?

Zero-risk bias is a preference for eliminating one risk completely over an alternative that would reduce more overall harm, when the relevant costs and consequences are otherwise comparable. The distinctive attraction is reaching zero, rather than achieving the larger total reduction.

A smaller risk removed, or more harm prevented?

Imagine a workshop with two sources of the same minor injury. For this simplified example, assume that the estimates are equally reliable, injuries have the same severity, and both groups of workers are treated equally. The workshop can afford one of two repairs, which cost the same:

• Repair A reduces expected injuries from one source from two per year to zero.
• Repair B reduces expected injuries from the other source from ten per year to five.

Before either repair, the two sources produce an expected total of twelve injuries per year. Repair A lowers that total to ten; repair B lowers it to seven. These are expected counts, not a promise about exactly how many injuries will occur. Under the stated assumptions, repair B prevents more harm.

Repair A nevertheless offers something B does not: one problem is completely gone. Choosing A just because its remaining risk is zero illustrates the bias. The relevant comparison is two injuries prevented versus five, not “safe” versus “unsafe.”

Where the research comes from

In a 1993 study, Jonathan Baron, Rajeev Gowda, and Howard Kunreuther used a mailed questionnaire about hazardous-waste decisions. Responses to simplified cases showed a preference for complete cleanup of some sites. These were hypothetical judgments, not observations of completed cleanup projects.

The concept is related to the certainty effect, in which certainty receives special weight. It is also more specific than risk aversion: preferring a safer choice does not, by itself, show an excessive preference for zero.

When eliminating a risk can be the better choice

Change the workshop example so that source A can cause a much more serious injury, or eliminating it also removes years of costly inspections. The original comparison no longer settles the decision. Severity, uncertainty, who bears the risk, and future costs can all change which repair is preferable.

Before calling a choice biased, put the options on a common basis: the same kind of harm, the same time period, and the full costs and consequences. Complete elimination may be justified. The problem is allowing the word “zero” to settle a comparison that those details should decide.

References

Baron, J., Gowda, R., & Kunreuther, H. (1993). Attitudes toward managing hazardous waste: What should be cleaned up and who should pay for it? Risk Analysis, 13(2), 183–192. https://doi.org/10.1111/j.1539-6924.1993.tb01068.x