After last week’s piece on uncertainty (and to answer some questions that were asked from it), I would like to add a bit more depth to some of the results I highlighted.

Unsurprisingly to anyone who has followed behavioural finance in any depth, it turns out us humans aren’t very good at assessing probability. One of the key findings from our work, was there is very little relationship between the latent personality trait of uncertainty acceptance with the types of gambles people are prepared to accept. This comes out through a series of questions where we ask people whether they would take the certain gain or varying uncertain, but larger, gains.

It would be expected that someone who is uncomfortable with uncertainty would be more likely to take the certain outcome (the riskless gain) rather than gamble on making more money. Unfortunately, this result doesn’t hold – even after we control for knowledge of financial concepts.

While – yet again – this may not come as especially surprising to those who have studied the field closely, when you look at the kinds of questions a standard risk profile questionnaire asks, you have to wonder what it is that they are attempting to measure.

As I mentioned last week, we see an assessment of tolerance for uncertainty as one of the primary goals in determining a risk profile. Ultimately, we are worried that people will regret their decision if it isn’t consistent with their inherent beliefs – something we see when “risk loving” investors sell their equities after short term falls. If people can’t stick with the retirement plan, then it is the fault of the plan designer rather than the person. At OnTrack, our aim is to build personalised retirement plans without the regret, to do it digitally and to do it cheaply for the mass market.

‍