Personalized savings reminders can help boost retirement savings
Tailored messages could encourage people to save more for retirement/ Article in Scientific Reports shows who responds to which message
Private retirement savings is an increasingly important topic in public debate. One popular idea is to phrase savings reminders in one of the two ways: either you emphasize what someone stands to lose if they don’t save; or you emphasize what they stand to gain if they do. To date, research into which of the two reminder messages is more effective in terms of influencing people’s efforts to save has produced conflicting results. A team of researchers from the University of Cologne, LMU Munich and Paderborn University has now demonstrated that people react very differently to the same message. Messages pointing to potential losses tend to have a greater impact on people whose jobs involve processing large amounts of information. Messages about potential gains, on the other hand, tend to resonate more with older people (aged 50 and over) and those who have already made numerous additional contributions.
The paper entitled ‘Causal machine learning reveals heterogeneous effects of loss versus gain framing on retirement savings’, published in the journal Scientific Reports, demonstrates the extent to which targeted behavioural interventions can increase private pension savings. In addition to Professor Dr Markus Weinmann from the University of Cologne, the project also involves Kevin Bösch and Professor Dr Oliver Müller from the University of Paderborn, and Professor Dr Stefan Feuerriegel from LMU Munich.
In 2021 and 2022, the researchers carried out two field experiments in collaboration with a European private pension provider, involving a total of around 41,000 customers, most of whom were based in Switzerland. They received a reminder via an app about the amount they could still save and claim as a tax deduction. Customers were randomly assigned a reminder framed either as a loss or a gain. The team used causal machine learning, a method that learns from experimental data for whom an intervention works and for whom it does not. Using data from the first experiment, the researchers estimated who responds to which message. They then tested this prediction on the customers in the second experiment.
On average, both messages have almost the same effect. However, there are significant underlying differences: among the quarter of customers who respond most strongly to the message referring to loss, the proportion of customers who make an additional contribution is 9.4 percentage points higher with the loss message rather than the gain message. In the quarter that responds most strongly to the message referring to gains, it is 8.4 percentage points lower with the loss message.
If every person received the message that was right for them, the researchers estimate that 51 per cent more customers would make an additional contribution (6.86 per cent instead of 4.53 per cent of those contacted). The average amount saved would rise by 18 per cent (2,442 instead of 2,062 euros per person who saves). The comparison is made using the alternative option of sending the message referring to loss to everyone. Weinmann, an economist from the University of Cologne, explains: “We show how much targeted behavioural interventions can do for private pension saving. If everyone received the reminder that resonated with them, around 50% more people would save extra for their retirement. On average, they would pay in almost a fifth more than they would if there were a uniform message for everyone. A differently worded message, sent to the right person, would be enough.”
Wissenschaftlicher Ansprechpartner:
Professor Dr Markus Weinmann
University of Cologne
markus.weinmann@uni-koeln.de
Originalpublikation:
https://www.nature.com/articles/s41598-026-76813-1
Ähnliche Pressemitteilungen im idw