The 'Asian disease' experiment that discovered the bias
In 1981, Daniel Kahneman and Amos Tversky presented their subjects with a hypothetical scenario: an outbreak of a disease threatening to kill 600 people, and two possible programs to fight it. One group was given the options framed in terms of gains: Program A would save 200 people for certain; Program B had a one-third chance of saving all 600 and a two-thirds chance of saving no one. Most people chose Program A, the safe option.
Another group was given the mathematically identical problem, but framed in terms of losses: Program C would certainly result in the death of 400 people; Program D had a one-third chance that no one would die and a two-thirds chance that all 600 would die. In this case, most people chose Program D, the risky option, even though it's exactly equivalent to Program B from the first group. The simple change in framing (lives saved versus lives lost) flipped the majority preference between the safe and the risky option.
How the financial industry uses framing to make the same number more appealing
The same mechanism shows up constantly in how financial products are communicated. A monthly-payment fee can be presented as a 'surcharge for splitting the payment' (a loss frame, which triggers rejection) or as a '5% discount for annual payment' (a gain frame relative to the single-payment option), even though both descriptions refer to the exact same price difference between two payment methods.
Similarly, a fund that has lost 10% can be described as having 'preserved 90% of its value' or as having 'lost a tenth of your money': the information is identical, but framing it in terms of what was preserved, rather than what was lost, tends to trigger a considerably less negative emotional reaction.
- 'Surcharge for monthly payment' (loss frame) versus 'discount for annual payment' (gain frame): same price difference.
- 'Preserved 90% of value' versus 'lost 10% of value': same figure, different emotional reaction.
- '99% historical success rate' versus '1% historical failure rate': same data, different perceived alarm.
A practical example: how return framing can mislead you
imagine a fund that rises 50% one year and falls 50% the next. Presented as an 'average annual return of 0%' (the arithmetic mean of +50% and -50%), it sounds neutral. But the real cumulative return after those two years is actually -25% (100 becomes 150, and that 150 then falls 50% to 75), not 0%. The 'average return' frame can make an outcome that was clearly negative for invested capital look neutral.
This gap between return frames isn't a minor technicality: understanding the difference between average return and real compounded return is exactly the same principle, applied in reverse, that we explain in our guide to compound interest, where the 'small annual percentage' frame can hide just how large the cumulative effect becomes over many years.
Why knowing this bias isn't enough to immunize you against it
One of the most uncomfortable findings about the framing effect is that it persists even in people who know Kahneman and Tversky's experiment perfectly well and explain it to others: knowing intellectually that framing shouldn't matter doesn't stop it from still shaping the immediate emotional reaction to information at the moment of deciding.
This makes the framing effect a fitting close for this series: like the nine other biases we've covered, it isn't a failure of intelligence or a lack of information, but a mental shortcut built into how the human brain processes decisions about gains, losses, and risk.
How to protect yourself from the framing effect
The most practical defense is a very specific habit: whenever you're presented with an important financial decision, reframe it yourself the opposite way (if it's presented in terms of gain, translate it into terms of loss, and vice versa) before deciding. If your decision changes depending on which frame you use yourself, that's a clear sign the frame, not the actual content of the information, is weighing more than it should.
And, closing the loop on this whole series: automating recurring financial decisions (saving, investing, portfolio rebalancing) remains the most robust defense against all ten biases we've covered, precisely because it removes the moment of emotional decision-making where each of them, from anchoring to framing, gets its chance to act.
Apply This Knowledge to Your Real Finances
eXcenda combines net worth, expenses, simulations, and macro context into a single app so you can move from theory to action.
Frequently Asked Questions
It's the cognitive bias by which the way the same information is presented, in terms of gains or losses, systematically changes the decision made, described by Kahneman and Tversky in their 1981 'Asian disease' experiment.
That, faced with the exact same mathematically identical problem, most people choose the safe option when it's framed in terms of lives saved, and the risky option when it's framed in terms of lives lost, even though both frames describe exactly the same outcome.
By presenting fees as 'discounts' instead of 'surcharges,' or negative results as 'value preserved' instead of 'value lost,' to trigger a less negative emotional reaction to the exact same number.
Because the arithmetic mean of a percentage gain and a percentage loss doesn't equal the real cumulative return: a +50% followed by a -50% gives a 0% average return but a real cumulative loss of 25% on the invested capital.
Not completely. Research shows this bias persists even in people who know the original experiment well, so it helps to rely on practical habits, like deliberately reframing information the opposite way before deciding.