Adjusted Risk Tolerance – Improving the understanding of the investor
Author Andre Correia
Date 16th November 2016
Risk Tolerance, or the willingness to take risk, has been the foundation of investor risk profiling methodologies for a decade. The FSA’s 2011 guidance says advisors’ must assess an investor’s overall willingness to take risk as well as their capacity to withstand any likely losses and not conflate the two. Risk Tolerance, when investing or in any other risky domain, is generally accepted to be a psychological trait, and as such is best assessed through psychometric methods. However, despite its proven usefulness, psychometrics has its detractors. They say the questions often lack precision, when in fact this is partly what makes them successful, and may find the ‘ordinal’ results challenging to interpret. Risk Tolerance assessments place people relative to each other, they don’t per se give absolute measures, this must be inferred from empirical calibration.
Moreover, because they reveal the investors underlying feelings about taking risk, Risk Tolerance assessments do not separate out the biases and distortions, unhelpful or otherwise, the investor is subject to. It begs the question; can Risk Tolerance assessments be enhanced to get closer to the investors true willingness to take risk?
Oxford Risk has recently completed an 18-month research project to answer this question, and the findings are very insightful into the overall strength and limitations of the traditional Risk Tolerance measures.
The study confirmed the strength of psychometric Risk Tolerance. It is a reliable, accurate and stable construct that places individuals on a relative scale of the willingness to take risk compared to other investors. As a starting point, this Raw Risk Tolerance is very valuable, particularly as the majority of investors find the typically short question sets easy to complete. However, the study confirmed that there are several respondent attributes, that systematically influence or bias the way they answer the Risk Tolerance questions. For example, a misplaced, or lack of confidence can skew the answers the investor may give. More usefully the degree of such biases can be linked to attributes and factors such as age, gender, knowledge and experience, and aspects of the investor’s financial personality and wellbeing, which can also be assessed and used to correct the biases. Correcting for these systematic biases results in a more accurate representation of their willingness to take risk, their Adjusted Risk Tolerance.
By way of example, let’s explore the influence of two such factors ‘Financial Understanding’, and ‘Future Outlook’ on Adjusted Risk Tolerance.
We use both objective and self-report questions to explore ‘Financial Understanding’. Where the objective measure differs from the self-report we can identify, for example, investors who may feel unjustly confident in their ability, but lack competence, leading to an unhelpful bias towards a higher willingness to take risk. Such a confidence bias can be corrected for in their Adjusted Risk Tolerance. Similarly an investor who lacks confidence, yet displays greater objective competence, might have their Risk Tolerance adjusted upwards.
We find investors with an irrational optimistic ‘Future Outlook’ will exhibit a systematic bias towards higher Risk Tolerance, whereas pessimistic individuals exhibit a systematic bias towards lower Risk Tolerance. Whilst this may seem intuitively true, optimism or pessimism, combined with the degree to which the investor is reward focused, or loss avoiding, will reveal a much more subtle adjustment to their Risk Tolerance.
Oxford Risk’s next generation of Risk Tolerance instruments will ensure that where they can be objectively revealed, biases and distortions will assessed and factored into a more reliable and accurate Adjusted Risk Tolerance measure. These adjustments may be small and incremental but in some cases this will be sufficient to move an investor who is near a Risk Tolerance category boundary, into the adjacent category. Moreover, the adjusted Risk Tolerance is more accurate because the Error of Measurement (EoM) is lower. (The EoM is the range around the calculated score that, with a certain level of confidence, contains the investor’s ‘true score’.) A lower EoM means less likelihood that an investor is placed in the wrong category.
Adjusting for the biases that influence Raw Risk Tolerance could make a difference to what will be appropriate advice, on investments suitable for the investor.
If you would like to know more about the next generation of Risk Tolerance assessments, and the implications of Adjusted Risk Tolerance, please get in touch.
Click on one of the links below to subscribe to the Oxford Risk podcast