The Power of Percentage: Quantitative Framing of Pension Income

Research Retrieved: February 2018
Retrieve the paper from Netspar


The way outcomes are framed has long been explored. For example, in treating the same illness, outcomes can be framed in terms of deaths or in terms of lives saved. This has shown to impact how patients choose their treatments (Tversky & Kahneman, 1981). The authors of this paper apply the concept of framing effects to pension communication. They ask whether framing the same replacement ratio in different ways yields different outcomes with respect to the satisfaction with this replacement ratio.

Key Findings

In a survey experiment with a representative sample of the Dutch population, the authors asked respondents to imagine that, given the respondent continues working until retirement, they will receive a replacement ratio of 0.5. This replacement ratio was either given as 1) the gross annual amount in euros (calculated with the respondent’s income), 2) the gross monthly amount in euros, 3) the replacement ratio as a percentage, or 4) the replacement ratio as a decimal. The authors find that respondents who were given the percentage replacement ratio are more likely to perceive their future pension income as insufficient. Respondents in the decimal replacement ratio group perceived the future income as more adequate. The relationship is robust, and holds when controlling for age, wealth, education, income, and saving attitudes.


Practical Relevance

Saving privately, and additionally, to first and second pillar pension becomes increasingly important. Before one can make the right decision on whether and how much to save, one needs to evaluate the pension promised by the state and/or one’s pension fund(s). If practitioners want to instruct a “wake-up call” so as to increase people’s additional savings, it seems that displaying the future income as a percentage replacement ratio is more effective. With such a frame, people are more likely to (correctly) perceive a gross replacement ratio of 50% as insufficient, compared to gross annual or yearly amounts and a decimal replacement ratio.