Risk profiling is an essential activity for wealth managers today but as they know only too well, many approaches come up short. This article looks at how this might be improved.

Risk-profiling of wealthy clients - in other words, trying to work out what their tolerance for risk is - is fraught with difficulty. It is sometimes said that it is often only during great stress, such as a severe market downturn, that a person’s real tolerance for risk shows itself. Considering that wealth managers must take care in this ever more regulated world to put clients into investments considered “suitable” for them, this issue becomes more pressing still. In this article, Bernard Del Rey, co-founder and chief executive of Capital Preferences, considers some of the challenges.

The author’s firm is a decision science and financial technology firm operating in the US, UK and Australia. Del Rey, a technology and financial services entrepreneur, is also founder of global strategic consultancy Capital Position Ventures and former head of strategic programmes and executive team member for what is now JP Morgan Private Client Services and chief marketing officer for JP Morgan Institutional Asset Management and Morgan Stanley Investment Management.

Current financial risk profiling for wealth clients is not up to the job. The financial crisis of 2008 revealed failures in the inadequacy and unscientific nature of client risk-profiling practices. These failings have yet to be addressed in any meaningful way. A 2011 review by the Financial Services Authority (now the Financial Conduct Authority) of 11 risk profiling tools found that nine had weaknesses which could lead to flawed outputs (the finalised guidance was called Assessing Suitability: Establishing the risk a customer is willing and able to take and making a suitable investment selection). This was the precursor to a warning in May 2014 that the FCA still had concerns about the way that advisors and – by extension – wealth managers use risk profiling tools.

New ways of approaching investment advice are needed in order for wealth managers and financial advisors to really understand their clients’ needs. In the coming years, many investors will once again find themselves in portfolios that only partially reflect their preferences and this discomfort will have them liquidating their accounts and firing or even suing their advisors. Given this and tightening regulatory scrutiny, many financial advisors find themselves between a rock and a hard place. On one hand, they are challenged by regulators to really understand their clients and their investment preferences so they can provide the right advice. On the other hand, regulators, resource challenged, continue to focus on enforcement over methodological improvement in client profiling techniques. 

The message to advisors should be a simple one – doing the minimum is not enough. The industry cannot wait for regulators to tell them how to conduct or protect their business.  Now is the time to bring more scientific rigour to how wealth managers diagnose and keep current with clients’ preferences.

You can tell me what you want…
The underlying weakness is largely in the widely adopted questionnaire-based "stated preference" method of risk profiling. It is unable to provide statistical confidence to support its classification of clients. What’s the value of a classification if the probability of its accuracy is unknown?  Moreover, stated preferences assumes that people can accurately state their preferences in complex arenas such as loss aversion or ambiguity aversion. This is unreasonable. Clients do not understand the definition of loss aversion, let alone share their self-assessment in a way that a financial services firm can actually use.

Because psychology-based methods are questionnaire-based, there is no mathematical or scientific proof through which the survey methods can be verified. Each of us could debate forever the questions and the order of the questions we might ask. In fact, some financial services firms have invented their own "methods" with no academic grounding or real world testing at all. In all of these cases, the explanatory power of stated preference methods can be neither proved right nor wrong because there is no mathematical theory or proof which they are trying to solve. 

These are deficiencies that cannot be ignored. Advisors often ignore the client profiling tools that are provided to them, because they teach them nothing about their clients and the clients find them annoying and pedantic. Only when there is a shock in the market do wealth managers and clients really discover the true level of risk-tolerance. This leads to lack of trust between the parties.

But actions speak louder than words
The FCA and other regulators worldwide are focusing on whether wealth managers can provide suitability of investments for clients. Therefore the need for a better understanding as to what clients’ investment beliefs are and how to match their expectations with investment preferences has never been greater. Recent academic research highlights the best way to understand investment preferences is not to ask a client, but to get them to play a game whose output gets them to reveal those preferences.

Why does the difference matter? Consider people’s preferences when it comes to dolphins. When asked, most people will state a strong affinity for dolphins, giving a stated preference score of around eight on a scale of one to 10 on how much they “like” dolphins.  However, when asked how much money, time, or anything else they have invested in dolphins, most report a score of zero - revealing that their actual or revealed preference for dolphins is very low.

Revealed preferences – bringing scientific precision to risk profiling
Revealed preferences is the new kid on the block of risk profiling. Its methodology is anchored in decades of economic research (see academic list below), but made accessible to the market today through advances in econometrics, game theory and computing power.

Revealed preferences is solving the risk profiling challenge through innovative approaches that give financial advisors unprecedented predictive power and confidence in understanding their clients’ preferences. For example, by playing a non-intrusive, simple "investment game" on a phone, tablet or laptop, taking only a few minutes and with no financial knowledge or advisor input required, a client will reveal detailed insights that inform the client relationship from the outset - insights that would ordinarily take many months or even years to develop. These insights are in the form of verifiable, scientific scores able to stand up to regulatory scrutiny. These scores bring very important precision to previously unanswerable questions such as “How loss adverse is this client?” or “How ambiguity adverse?” and “How much more does this client value a pound today over a pound in a month?”

Financial services providers implementing this kind of revealed preferences approach can use these scores for more than risk profiling:

- Marketing managers wanting increased effectiveness in marketing campaigns and higher customer retention need tools and metrics that lead to better client segmentation and higher marketing return on investment;

- Scores from the investment game can be linked with a client relationship management system to identify individual clients’ behaviour, provide more precise client segmentation and support strategic marketing plans;

- Risk managers needing a scientific method for client profiling that addresses growing scrutiny in regulatory requirements can use the scores to support consistent risk profiling across channels;

- Chief investment officers wanting consistent implementation of asset allocation strategies that meet each client’s need require a tool that supports client-specific customisation within target allocation ranges across a range of factors;

- Revealed preferences-based methodologies can provide custom portfolios, or robust grouping in standard ones – linking asset allocations with individual client needs.

Where do we go from here? The future is now
Regulators, financial advisors and wealth managers need to begin the hard job of putting advice on a scientific grounding. If advice givers are going to defend their model and improve clients’ financial future, they need to understand more about their clients and their preferences.  The revealed preferences approach offers a clear break from risk profiling’s unscientific past. Regulators and advice givers should grab the opportunity with both hands.

Academic References
Choi S., R. Fisman, D. Gale and S. Kariv (2007) "Revealing Preferences Graphically: An Old Method Gets a New Tool Kit," American Economic Review, 97, pp. 153-158.
Choi S., R. Fisman, D. Gale and S. Kariv (2007) "Consistency and Heterogeneity of Individual Behavior under Uncertainty." American Economic Review, 97, pp. 1921-1938.
Choi S., S. Kariv, W. Müller and D. Silverman (2014) "Who Is (More) Rational?" American Economic Review, 104, pp. 1518–1550.