The Ombudsman highlights problems identifying and matching clients’ risk tolerance
Consumer complaints about investments have nearly doubled in the past year, according to the Financial Ombudsman Service’s annual review. Overall investment and pension complaints (now including mortgage endowment cases) jumped to 22,265 from 12,787 in 2007/08. Commenting on the report, Citywire’s ‘New Model Adviser’ forum singled out that ‘while the ombudsman recognised under performance of investments was a factor in some of these complaints, it said poor stock market conditions had exposed poor advice. In particular it identified a trend in complaints where financial advisers had not sufficiently considered clients’ tolerance to risk.’
Having taken a complaint against one of the banks to the FOS, I can see that there is a significant risk of a lottery effect in the adjudication of blame in this area, simply because the typical process of discovering risk attitudes is so unscientific, unquantified, jargon-filled and partial. In my book I referred to it as ‘woolly flannel’. Six years later, nothing has changed. This applies equally, by the way, to banks, IFAs and private-client portfolio managers. You can read a slightly expanded version of the comment I posted on the Citywire site, for the benefit of our peers, suggesting more quantification would bring clarity to both parties and help avoid disputes.
My firm has never had a complaint. We operate as both planners and discretionary managers and we are far from risk averse in our advice. ‘Covering your backside’ usually diminishes the value of advice to the client. I think part of our advantage lies in quantification: using numbers to differentiate and describe rather than woolly flannel.
I believe many advisers think that using financial modelling exposes them to additional risk simply because the information provided to support advice is so much more explicit. I disagree. It is woolly flannel, whether in conversations about risk attitudes or in product descriptions, that is vulnerable to multiple interpretation and hindsight bias.
By ensuring there is greater clarity about risks, such as when quantifying expected probability ranges for both short-term performance and longer-term outcomes, clients are likely i) to exhibit true risk preferences and ii) to avoid regret and discrimination when things turn out badly.
Being this explicit about the realistic probabililties can put clients off risk taking. But that can itself be balanced by being equally explicit about the costs of avoiding risk (which are usually market-priced).
Quantification, btw, requires its own legal boiler plate clauses in agreements, because models are still only models.
I have taken on pro bono an FOS complaint against Lloyds TSB. I was surprised and pleased to get my hands on internal documentation that both controls and records their sales process. Whilst I have no doubt that management incentives encouraged bad advice, so also do technical weaknesses in those processes, even though they are clearly designed to reduce the chance of successful post-sale complaints.
A common source of the weaknesses is ignorance about the true return probabilities in the typical products they promote, given their underlying asset allocation. There is no excuse for this given the wealth of historical data available to them as well as access to explanatory theories.
This lack of technical knowledge, which would be helped by modellling, is unfortunately widespread in the IFA community too. It is evidenced by the typical statements about risk and time horizons and by reliance on diversification to manage risks rather than limited exposure to those risks.
As an industry we have much to learn about investment returns before we can be sure of avoiding giving bad advice.