Tuesday, 5 April 2016

The risks and benefits of automated financial advice

In December 2015 the three European Supervisory Authorities (ESAs) (the European Banking Authority, the European Securities and Markets Authority, and the European Insurance and Occupational Pensions Authority) issued a joint Discussion paper on automation in financial advice.

The ESAs have recognized that with increasing digitization of financial services, more and more financial institutions offer automated financial advice to their customers (also called 'robo-advice').Thus the Discussion paper is aimed at assessing what (if any) regulatory/supervisory action is required to enable consumers and firms to take advantage of the benefits and to mitigate the potential risks, of automated advice. The Discussion paper explains the concept of automated advice, and highlights the possible risks and benefits of this innovation.

Automated advice means that a recommendation to buy or sell financial products is generated by automated tools (typically websites using algorithms or decision trees) without (or with very limited) human intervention.

The ESAs believe that automated financial advice may be beneficial for consumers in terms of providing easy access to a quality service (free from behavioural biases, human error or poor judgment and reliant on a large volume of complex data) at a considerably lesser cost than traditional advice. However, they also recognize that automated advice may carry a great deal of danger. Importantly, consumers may find it more difficult to understand automated advice than traditional advice without a human interaction and being able to ask questions and seek clarifications; consumers may misunderstand the nature of the advice (they may receive general advice but believe that they have received personalized advice); consumers may end up with unsuitable advice because they do not understand how information is used by the automated tool and enter incorrect or incomplete data, or because of a failure in the automated tool itself. Finally, automated financial advice, as with any digital service, raises data protection issues.

The consultation period ended in March 2016. The Discussion paper generated great interest among the stakeholders and the ESAs have received many responses. BEUC welcomed the ESAs interest in automated financial advice, and considered the Discussion paper to be a well balanced discussion of the risks and benefits of automated advice. It has however highlighted that protecting consumers in this area will require new approaches, and that market outcomes may largely depend on the quality of algorithms used for guiding consumers through the advice process.

What do you think? Is automated advice more likely to benefit or to harm consumers?