How psychology works when funds should assess their retirees


If you are wanting to know someone’s risk tolerances, don’t ask him or her. Their ‘stated preferences’ will be different from their ‘revealed preferences’. The application of behavioural finance to risk has come a long way. Milliman seminars last week explained how.

According to Professor Shachar Kariv of the University of California, Berkley, who addressed Milliman’s breakfast seminars in Sydney and Melbourne, as well as an IMCA lunch in Sydney, the traditional surveys which try to assess risk tolerance have no basis in science.

“There is no mathematical theory behind these surveys,” he told the Sydney Milliman event. “There is very little economics in the behavioural economics of this type. The methods being used to understand risk preferences are not the right methods.”

This reflected the difference between ‘stated’ preferences and ‘perceived’ preferences. It’s the ‘perceived’ preferences, garnered from a different set of one-on-one questions and real-life observations, compared with ‘stated preferences, garnered usually from questionnaires, which actually matter.

Milliman, a big global actuarial firm which specialises in risk and which has also developed a range of strategies for funds to build retirement products, argues that retirees need more bespoke investment products and strategies, which technology can now help to provide in a cost-effective fashion.

Craig McCulloch, a Milliman principal, said that this was about providing interactive and on-demand content for members. “It’s about engagement and education in a format which is fit for various segments,” he said. “It’s about tailored communications… We are seeing this trend play out very quickly.”

Kariv said: “At the end of every survey they do some weighted averages of the answers to do with risk tolerance and they end up with the number ‘6’. This means nothing except that it is more than ‘5’ and less than ‘7’. Everything needs to be done within a certain range of probability…”

The University of California, Berkley (there are several campuses) is considered the “Mecca” of behavioural economics, Kariv said. He is leaving his current Professor of Economics post soon for another one elsewhere.

What he and his colleagues have been studying is investor behaviour under uncertainty. They have laid the groundwork for a new field of research, taking behavioural finance to a new level.

A financial advisor, for instance, needs to know three different terms: risk requirements; risk capacity; and, risk tolerance.

‘Revealed preferences’ have to do with putting someone’s risk tolerance under the microscope. For instance, Kariv said, there was a difference between ‘risk tolerance’ and ‘loss tolerance’. There was also a difference between ‘risk aversion’ and ‘ambiguity aversion’.

Ambiguity was where the investor did not know the probabilities to do with an investment decision, he said. “People tend to shy away from stocks not because they are risky but because they are ambiguous,” he said. “They want stocks which are close to them and that they understand.”

Jeff Gebler, a senior consultant at Milliman, said that when you looked at spending in retirement, averages could be misleading. People tended to spend less as they got older, and the range between the top and bottom spenders narrowed.

Also, he said, the “basket” of goods and services on which they spent was not necessarily reflected among the broader community. They tended to travel less, for instance, and spend less on entertainment. “The CPI might not be the best benchmark for setting spending targets in retirement.

Milliman has a service known as “Expectations and Spending Profiles” (ESP) which is a part of its bigger-picture assessment of retirement strategies. It provides an analysis of differences in expected spending behaviours and also the issues likely to face super funds as a result.