string(28) "www.ifamagazine.com/article/" Next Page

The risk of risk-profiling – Keith Robertson sounds a warning bell

  • By Sue Whitbread

The assessment of risk is fundamental to advisers and planners in their quest to ensure that clients benefit from effective, appropriate portfolio construction and asset allocation. In this article, the first of a two-part series for IFA Magazine, Keith Robertson casts his analytical eye over the weaknesses of common risk assessment techniques and sounds a warning bell to advisers as to their limitations. 

Probably the biggest regulatory and legal risk for financial advisers today is their use of Risk Profiling-cum-Asset Allocation (RiPAA), which has become ubiquitous and the overwhelmingly predominant investment process.  However, it is only the latest fad used by intermediaries.  Twenty-five years ago a with-profits fund was the vade-mecum for all occasions, followed by mean-variance optimisation, stochastic modelling, core and satellite, rebalanced 60:40 passive investing, multi asset, multi-manager, model portfolios, risk-adjusted and risk-targeted portfolios, etc. – and now, our very latest model: RiPPA with systematic rebalancing!  Its uncritical and unthinking acceptance does not speak well of our sector.

Science or magic?

RiPPA is often touted as being ‘scientific’ which it is not; it is MAGIC!  Consider.  You can choose any one of dozens of risk-profiling questionnaires.  Each has 5 to 25 multiple-choice questions, from which one option must be chosen in answer to each question, even if none is the answer clients want to give.  The questionnaire is marked somehow, and the client informed that, Hey Presto! this gives the perfect asset allocation and portfolio for you to three decimal places.

This basic process is held out to work for any client, irrespective of gender, age, wealth, socio-economic background or personal circumstances, educational level, financial knowledge and experience, IQ or other measure of cognitive ability.  It supposedly works equally well for a 34-year old entrepreneur who is a member of Mensa as for a recently-widowed 84-year old retired teacher whose husband had managed all their finances, as for a 24-year old single mother with severe learning difficulties who has just inherited a million.  Now that’s magic!

The driver of this mini-industry is regulatory.  The key policy statement reads, “Ascertaining a private customer’s true attitude to risk is critical for any adviser in assessing suitability and making an investment recommendation [FSA 2008, 2011].”  While this appears to be clear, even directive, logical analysis shows such a process to be absurd, impractical and impracticable.  This does not mean anyone’s anxieties about something going wrong and losing money is unimportant; it is of supreme importance.  But the notion that such concerns can be reliably measured quantitatively at the outset to derive a suitable asset allocation is idiotic.  Nobody’s attitude to risk (ATR) is fixed and stable; it is context-dependent and liable to continuing change over time.  Investment risk is not fixed and stable, resident permanently within a product, so that financial planners, like pharmacists, can pick suitable remedies off the shelf, confident that investments will be entirely predictable.

What is supposed to be measured? 

Questionnaire providers are touchy about this.  All claim to measure a specific psychological trait: risk attitude (ATR), risk aversion, risk tolerance, risk appetite, etc.  The belief that one questionnaire, largely indistinguishable from all the others, should be able to accurately measure one of half-a-dozen discrete psychological traits for purposes of creating a suitable investment portfolio, should itself be a warning.  The notion that a range of traits can each be exclusively, specifically and scientifically measured is hogwash.  All use similar questions and, without the manufacturer of trait being identified, it is pretty much impossible to differentiate among them.

Each claims to measure some investment risk-related quantity resident as a psychological artefact within each client: it is the questionnaire’s raison d’être.  Advisers should note the regulator’s requirement to ascertain your client’s ‘true attitude to risk’.  A good start is to ask the client what he understands ‘risk’ to be.  Unless you both understand ‘risk’ to be exactly the same thing, neither of you will know what the other is really talking about.  Then ask your questionnaire supplier to define what ‘risk’ means when used within their questionnaire.  Unless all three parties understand ‘risk’ to mean the same thing, it will be a case of the blind leading the blind and heading for serious trouble.

Next, ask your questionnaire supplier to define what ‘risk tolerance’, ‘risk aversion’ or whatever is the trait being measured.  If you want to measure anything, you have to be able to define it.  If you cannot define it, how can you be certain what you are measuring?  The RiPAA manufacturer must also be able to explain how their risk factor (e.g. ‘risk tolerance’) can be distinguished from all the others (‘risk aversion’, ‘risk appetite’, etc.).  The next logical step is to require the provider to explain, question by question, how each question can be proven to be measuring only, say ‘risk attitude’, and not accidentally measuring ‘risk tolerance’, or ‘risk appetite’.  If questions turn out to be too general or generic and capable of measuring other traits, how can the manufacturer claim to be reliably measuring only their chosen trait?

These questions will clarify whether your risk-profiling provider has an intellectually robust proposition or is a conman.  They go to the heart of the unfairness inherent in using such questionnaires and the unnecessary danger caused to clients by misleading them into believing they are participating in an honest, accurate and reliable process.

Petitio principia

In logic, petitio principia is about the most egregious fallacy that can be committed.  It is a type of circular reasoning where you assume the conclusion in the starting assumption; it seeks to prove a proposition when the premiss itself requires proof.  In risk profiling, the fact that a provider claims the questionnaire will measure, say ‘risk appetite’, and it generates a result called the client’s ‘risk appetite’, and everyone acts as if they believed that is what has been measured, does not in itself mean that ‘risk appetite’ has been measured, or indeed that anything useful at all has been measured.  Unless the provider has met the earlier tests of being able to define ‘risk appetite’, ‘risk tolerance’ or whatever, and demonstrate its questionnaire will measure only that quality in a valid, accurate, precise, reliable and robust way, there is no reason to believe that ‘risk appetite’ or ‘risk tolerance’ etc. even exist in a way that can be measured.  One would need to be a very brave adviser to buy into risk-profiling and use any questionnaire which fails these basic tests.  Even if the regulator continues to equivocate about RiPAA, it is likely that the courts will decide for them at some point.

 

Context-dependent

Financial planners know investment advice is, above all, a practical skill.  They are well aware a client’s propensity to bear the risk of loss is variable, and dependent on subjective factors prevailing at the time.  Whatever ‘risk’ is, advisers practising in 2000-02 and 2007-09 are painfully aware that clients who suffered losses in those crashes wished they had had less.  It was not that clients had not been told that financial markets are risky and severe falls theoretically possible; it was that they did not realise they would feel so badly and regretful about loss when it happened.

In other words, feelings about risk are context-dependent.  Risk questionnaire designers never take context into account, because it makes life too difficult.  Examining perhaps 200 questions from commonly-used questionnaires reveals they are designed for a crude algorithm to fit answer scores against volatility of assets that are, in turn, used to populate a set of predetermined model portfolios.  It is this monocausal interpretation of risk, totally described solely by subjectively selected statistical volatility, combined with denial of the contextual reality of risk, that contributes to the entire RiPAA process being useless for managing risk.

We must ask a fundamental question: what should we be trying to reveal by risk-profiling a client?  The answer, like ‘risk’ itself, is multi-faceted and dynamic.  A client’s willingness, indeed ability, to bear the consequences of something bad happening is never a fixed quantity.  It is labile, changing over time according to circumstances (both external facts, and internal psychology) in the future.  What you might feel today about hypothetically losing 20% of your wealth, completing an IFA’s questionnaire after a good meal and news of promotion in your firm, is likely to be very different from what you would feel two years from now when the markets have crashed, recession started, and you have been sacked.  Which of these two scenarios matters?  How you feel right now, or how you are likely to feel when a severe loss coincides with a personal disaster?

If questionnaires were constructed to try to uncover how a client might feel in the future if an investment reverse came coincident with one of the key life events (e.g. loss of job or income, divorce, death of a loved family member, diagnosis of a severe or terminal illness) then they would provide useful information, allowing time to reflect before an unthinking drive to get the person’s cash invested as soon as possible.  It is someone else’s money being risked.  An adviser’s job is to assist non-professionals make fully-informed critical decisions, not to push them into a RiPAA sausage-machine that magically spews out a portfolio, job done, so the next client can be put through an identical process.

About Keith Robertson

Keith Robertson spent over twenty years as a practising fee-charging financial planner and investment manager, and was one of the highest qualified practitioners in the UK.  Active in both the CISI and PFS-CII, he reviewed and co-authored professional exam course-books, was London’s inaugural Chartered Champion, and continues to sit on CISI masters’ level exam panels and forum committees.  Since 2016 he has focused on consultancy and training, particularly investment strategy and risk.  He is a fierce critic of current orthodox systematised and quantitative approaches to investment, concerned that advisers are not taught critical thinking to analyse its weaknesses, leaving clients in no position to understand the risks they are being exposed to.