When Jorgo obtained his PhD in 2023, one specific part of his thesis became particularly relevant. With the transition to a new system fast approaching – it must be completed by 2028 – parties within the pension sector want to know everything there is to know about risk, time and sustainability preferences. ‘A lot of research is currently being done on this. I have also continued to work on it myself. The pension sector is open to experimentation, because there is still a lot to learn.’
Major challenge
We are currently in a transitional phase, Jorgo explains. In the new system, pensions will be more in line with the preferences of participants. ‘We are moving from defined benefit agreements to defined contribution agreements. Previously, there were collective schemes where the outcome was fixed to a certain extent. Now the contribution is fixed, but the final pension outcome depends on factors such as investment results and how participants want to deal with uncertainty. In that sense, pensions are becoming more tailor-made.’
There is a difference between pension accrual through insurers and through funds. ‘With insurers, the emphasis is more on individual choices. Pension funds mainly work with collective solutions, whereby participants are divided into cohorts, for example based on age. The challenge is to align these collective choices as closely as possible with what participants consider important.’
Stability or significant fluctuations?
In order to translate participants' preferences into an appropriate pension, a large, representative sample will be taken periodically. This will measure risk preferences: how someone perceives the trade-off between certainty and uncertainty in pension outcomes. For example, does someone opt for a more stable pension or, on the contrary, for more fluctuations with a potentially higher expected pension ? The trade-off between now and later is also important. ‘People who live from day to day are, on average, less attached to future outcomes. Our research with pension fund participants shows that they are relatively more likely to stick to a standard choice that pension funds devise for them.’
‘Not immediately apparent from your expression’
Measuring preferences is not easy. ‘Your preferences are immediately apparent from your expression. In addition, they are partly dependent on the context,’ says Jorgo. All kinds of factors play a role. Age, for example: older people are on average often more risk-averse than younger people. Emotion can also have an influence. ‘If a participant is anxious at the time of measurement, this can affect their preferences.’ Furthermore, major macroeconomic events such as financial crises, pandemics or wars can cause preferences to shift. Another factor is whether a participant sufficiently understands what exactly is at stake. 'Sometimes it is difficult to question participants with language difficulties or lower literacy levels. We are talking about large sums of money and the possibility that something will happen, which can be quite abstract.'
Science and practice
Goossens works one day a week at pension administrator APG as an academic advisor. This gives him access to large-scale, anonymised data and insight into participants' actual pension choices. ‘I can develop, test and apply new measurement methods within the context of APG. I can then link the results to data on actual pension choices recorded in the datasets. This means it doesn't remain limited to hypothetical choices.’ The knowledge this yields is valuable for both science and APG. ‘And indirectly, it also benefits participants greatly,’ says Goossens. ‘We are working on their pensions behind the scenes.’
The tricky thing is that one method often leads to a different measured risk preference than another. There is still a lot to discover in this area. It is a large field with a lot of scientific literature, but accurate measurement remains a challenge.
Serious game
The search for good measurement methods is a challenging task. Jorgo and his colleagues developed a serious game aimed at participants with lower literacy levels. ‘The idea was to make it more fun and thus hold their attention for longer.’ In the game, you are the captain of a boat. You sail to islands, each of which represents a specific pension. ‘It was great fun to do, but in the end we found little difference compared to standard questionnaires. We did discover that participants in the game did not systematically take more risks, which you might expect based on the game design.’
Jorgo continues to experiment with the form and content of questionnaires. For example, he worked with APG on a method where you can choose between two pensions for each question, which makes the picture of their preferences increasingly specific. ‘The tricky thing is that one method often leads to a different measured risk preference than another. There is still a lot to discover in this area. It is a large field with a lot of scientific literature, but accurate measurement remains a challenge.’
Refinement
APG is very pleased with the collaboration with Radboud University, says Onno Steenbeek, Director of Strategic Portfolio Advice at APG and Associate Professor of Financial Economics at Erasmus University. ‘Asking participants about their risk preferences is a complex and, at the same time, important step for funds in developing a financial strategy. As an expert in this field, Jorgo has helped APG develop a workable approach. In the coming years, we will continue to refine this approach together.’
Text: Willem Claassen
Would you also like to collaborate with researchers at Radboud University? Please contact externerelaties.fm [at] ru.nl (externerelaties[dot]fm[at]ru[dot]nl).