INVERCO (Asociación de Instituciones de Inversión Colectiva y Fondos de Pensiones)

INVERCO mainly agree with the characteristics of the automated financial advice set out in the Discussion Paper.

However, the concept of “financial advice”, with a huge impact for the entities subject to MiFID, has been broadened in this Paper, identifying as one of the three main characteristics of automated financial advice tools that the output of the tools is, or is perceived to be financial advice by the consumers. The latter could be considered, among others as below explained, as a characteristic of financial advice in general, but not in particular for the automated financial advice.

Of course, it can be possible to provide advice with an automated tool, but we should only consider that a financial advice is provided in the securities sector, whether automated or not, if it is provided in accordance with MiFID rules and in relation to MiFID products and instruments. For this purpose, it is crucial to take into account the definition of investment advice of the CESR’ document “Q&A: understanding the definition of advice under MiFID” (CESR/10/-293) according to which there are five tests (1) that all of them have to be met for a service to be considered as investment advice . One of these five test is the investor’s view of whether advice is being given. In this regard, CESR says “CESR recognizes that a particular client’s understanding of the nature of the service he is receiving will not always be accurate. For this reason, whether or not a particular client feels that he is receiving a personal recommendation will not determine, on its own, whether or not investment advice is actually being given”. Therefore, there will be a financial automated service in the securities sector only if the five tests are met and, among them, the perception of the consumer shall be assessed in accordance with the abovementioned CESR’s understanding and, of course, in no event as a main characteristic of the automated financial advice.

On the other hand, the use of automated tools can also be used, for example, for the provision of a “guided sale of financial instruments”, that is, as tools designed to help consumers to take investment decisions but without consideration of their personal circumstances. This would be an extraordinary tool for clients but would not be understood, and hence should not be presented as such, as a financial advice tool.

The same would be the case of instruments which don’t fall under MIFID rules, such as, for example, mortgages, personal loans, pension funds or insurance products. The advice for those type of products shall be analyzed taking into account the relevant rules under the banking, pension funds or insurance legislation but cannot be considered as financial advice in the securities sector.

Our understanding is that the terminology is important, especially considering the regulatory and legal consequences for the entities providing the relevant financial services.

(1) The five cumulative tests are the following:
1. The service being offered constitutes a recommendation
2. The recommendation is provided in relation to one or more financial instruments
3. The recommendation is presented as suitable or based on a consideration of the person’s circumstances
4. The recommendation is issued otherwise than exclusively through distribution channels or to the public
5. The recommendation is made to a person in his capacity of investor or potential investor o as an agent for an investor or potential investors.
Although this kind of tools are presented with almost a total lack of human intervention, it appears that in practice they would normally be designed with the possibility to permanent access to an operator by chat, mail or telephone, to help the customer out all throughout the process.
We are aware of a very few Spanish financial advisors (EAFI) that provide investment advice exclusively through automated advice tools in the securities sector. Unfortunately, we do not have the operating details on those automated advice tools.

All of them have been authorized by the Spanish securities authorities very recently for these purpose and it seems that they provide the investment advice service limited to investment funds products.

On the other hand, we are aware that several Spanish credit entities are currently working on the possibility of offering several financial services, not only investment advice, on a fully integrated digital model.

Finally, the use of automated tools to provide information, comparison websites, calculates is common in Spain.
NOT APPLICABLE (we are an association and we do not provide financial services)
We understand that the offering and developing of automated financial advice tools in the securities sector is possible. However, such tools should be construed taking into account all relevant legislation in relation to the services provided and the products and instruments recommended.

As explained in Question 1 above, the fact that the service provided can be considered as financial advice in so far as the consumer perceives it so, could be a major barrier, since the entities could be reluctant to assume the risks and consequences of such assumption when their intention is not to provide an investment advice service, particularly when they are relying on a third party design of the tool.

Likewise, our understanding is that the automated tools could be used to provide consumers with a global advice, that is, an advice comprising products from different sectors simultaneously (MiFID instruments, bank deposits, pension plans, insurance contracts) to the extent that those products can be offered by financial entities as interchangeable and/or complementary products. The fact that each sector is subject to different rules when providing advice on them can act as a barrier preventing the automated tools to provide such global advice (which at the end of the day can benefit the consumers). Therefore a way to minimize or to remove this barrier is to align the regulatory and legal rules applicable to the advice of each of the sectors (securities, banking and insurance and pension).
We consider that the potential benefits are, in general, accurately described.

Considering the growth of new technologies and the globalization of the access to Internet from any electronic devices, an automated tool will allow the quick and easy access to added value services for a broader range of customers. The provision of such services by a high number of entities will grant access to a wider offer of products to the customers, at, in principle, for a lower cost.
(i) The clients’ portfolios might be adjusted on a dynamic ongoing basis, in accordance with market conditions and any other relevant factors, provided that the automated tools have been fed with the relevant data for this purpose.

(ii) The client might always have access to his/her updated position at market prices, provided that the automated tools have been fed with the relevant data for this purpose.

(iii) The main benefit is to have the possibility to provide to all kind of clients the access to advisory services which, without such a tool, would probably be totally out of reach of many people, considering the costs they would imply when rebates are totally banned or considerably restricted.

Taking into account the development of the regulation on inducements, not promoting this kind of alternatives would imply the actual expulsion of many clients with lower assets under management from financial advice services.

On execution only tools, an additional benefit would be the possibility to follow up and review the former investment decisions and the grounds on which they were adopted.
In general, we do not see any relevant differences, although the products are different and the applicable regulatory and legal rules for each of them.

As explained in Question 5 above, a way to minimize or remove the barrier that may imply the fact that each of the sectors are subject to different regulatory and legal rules, is to align them for all the sectors.
We consider that the potential benefits are, in general, accurately described. The provision of advice through automated tools is underway and is able to eliminate subjective interpretations and human mistakes. Costs are likely to be lower.

However, the impact on costs should be specially analyzed considering the following:

(i) The costs of development of the tools and the granting of permanent online access to reliable sources of data would presumably be high.
(ii) The possibility of having access to human contact would probably always be kept.
(iii) Qualified personnel should probably be needed to permanently check and verify the accuracy of the outcome of the tool, the algorithms used, etc.
(iv) Despite the fact that potential human errors could be reduced, there might be other costs derived from the misuse of the tool, any flaws on the use of the tool or any other issues that could imply civil liability.
The use of automated tools would allow the entities to maintain accurate and up to date records of all the processes which would enable them to reconstruct and track all steps in the transactions, the communications with the clients and provide complete evidence of the same in relation to any claims, administrative proceedings or law suits to the relevant client, supervisor or court.
In general, we do not see any relevant differences, although the products are different and the applicable regulatory rules for each of them.

As explained in Question 5 above, a way to minimize or remove the barrier that may imply the fact that each of the sectors are subject to different regulatory and legal rules, is to align them for all the sectors.
We would like to make a few observations to the risks described in the Discussion Paper:

(i) Risks to have limited access to information and/or limited ability to process that information.

This risk would normally be mitigated by the possibility to have access to a person who will provide assistance by chat, email or telephone.

It would be of importance to place proper due diligence on the algorithm used by the automated tools in order to avoid negative outcomes in the financial advice provided.

(ii) R2. Risk to receive unsuitable advice as a result of not being made aware how information they input is used by the automated tool.

This risk is not specific to the provision of automated financial advice, since this could also happen on a face to face advice. Besides, the fact that a customer does not understand how the information is processed, this does not make the recommendation invalid. What is important here is that the advice or recommendation is valid according to the responses provided by the customer to the questions asked by the automated tool.

Customers request professional assistance since they do not have the knowledge and information to take their decisions. To ask the customer to understand how the information is used by the automated tool is requesting them a knowledge far beyond normal possibilities/conditions.

(iii) R3. In relation to the risks described under numbers 57, 58 and 59 are specifically addressed in the applicable legislation to MiFID services. Therefore, in those cases, there are not any different risks to the ones the clients have when they are receiving face to face advice, and thus the existing rules, and not a specific ones, should be applied to these risks arising on automated financial advice.

(iv) R4. As regards tailored recommendations, the risk would be mitigated if the specific service being provided to clients is clearly explained to them. Again, as above explained, it should not be presented as financial advice in the securities sector on a general basis if it does not meet the MiFID requirements for this purpose.

When we are talking about pure investment advice or even guided sale of financial products, it is true that for customers that provide the very same answers to the questions, the outcome of the proposal or recommendation, as the case may be, may be equivalent or even identical. However, that does not imply that it is not tailored to that client. Actually, if the customer changes any of the data, the recommendation will vary too.

(v) R5.Consumers do not understand who is providing advice (...)

The risks described in relation to the counterparties providing the services will probably be mitigated since the customer will always be able to identify the entity providing the financial service. This is one of the highest risks for the financial entity itself, especially when it does not develop the tools itself but instead has agreements in place with one or several entities for this purpose.

For the client purposes, it will be the same entity who will be capturing the data and then providing the recommendations, despite the existence of services agreements behind.

(vi) R6. Risks related to the use of personal data.

These risks are not different from the ones which may arise when providing such data by the customer on a face to face relationship. We understand that they are sufficiently addressed in the applicable legislation about data protection, which establishes the need to inform and, in many cases, obtain the customer’s authorisation for the use of the personal data provided. Hence, this is not a specific risk of the automated provision of services.

(vii) R7. Risk of making unsuitable decisions because of limitations or assumptions within the tool.

This is not a specific risk of the automated provision of the service and, if any, it would normally be lower than the risk inherent to face to face relations, always subject to subjective interpretations.

(viii) R10. Consumers make unsuitable decisions because the tool facilitates them to move too quickly throughout the process.

This risk is not specific or different from the face to face provision of the services. Actually to the contrary, since the customer may take time to answer the questions, leave them and return to them later or even modify them. We don’t consider this as a specific risk of the automated service.

(ix) R11. Consumers lack of motivation to act on advice where such tools do not facilitate and end-to-end process.

This is a very remote risk, taking into account the following:

(a) The client who access this kind of tools is looking for the service and, therefore, it is unlikely that he/she loses motivation as a consequence of that.
(b) Many tools will be designed to facilitate an end-to-end process or, otherwise, it will inform about such limitation beforehand.

(xii) R12. Consumers lose out as a result of automated advice tools being based on similar algorithms.

Automated tools normally use their own algorithms, differentiated from others. In addition, if there were a concentration of recommendations, the recommended product would likely increase its price and, therefore, would no longer be competitive and the algorithm itself would probably not recommend it anymore.

(xiii) R13. Consumers may no longer be given the opportunity to access any human advice.

As it has already been explained, as a general rule the customers will always have access to personal help in addition to the automated tool.

The tools will probably be able to be used also by financial advisers, in order to help the clients all throughout the process or at the end of it.

(xiv) Some of the risks which have been presented as customers’ risks are more applicable to the entities providing the services, as below explained.
We are not aware of any significant risks not included in the Discussion Paper.
In general, we do not see any relevant differences, although the products are different and the applicable regulatory and legal rules for each of them.
(xv) R14. Financial institutions may be exposed to litigation and subsequent reputational risk due to faulty automation.

The risk of litigation is always present in the provision of financial services. It is true that the possible flaws of the automated tool introduce an additional fact for litigation purposes while, on the other hand, the potential source of litigation due to human errors existing on a face to face relationship are mitigated.

The most relevant risks, in our view, for the entities providing the service is in case of financial entities don’t develop the tool themselves and/or don’t provide the information themselves, the legal disputes concerning the allocation of liability, especially when the entities providing the different services are located in different jurisdictions. The entities with whom the customers have the relationship will, in addition, face the direct claims of the customers and are likely to assume any potential liability, despite their right to claim the entity or entities liable for the damages.

(xvi) R15. Automated advice is likely to live together with human advisors, at least during the early stages of the automated financial advice development. However, we don’t think this is a real risk, both will have its own clients. The customers who look for an automated advice are normally not very keen to personal contact with their financial entities and, to the contrary, the people who value personal advice are less likely to use automated tools.

In addition to that, automated advice will probably be cheaper, considering that it requires less personnel involvement and reduces manual processes associated to such service. This will probably encourage the use of the automated tools by certain clients.

(xvii) R16. Legal disputes.

This risk has already been addressed under (xv) above. We understand, however, that this risk would be mitigated by the relevant legal agreements entered into between the different actors.
We are not aware of any significant risks not included in the Discussion Paper.
In general, we do not see any relevant differences, although the products are different and the applicable regulatory and legal rules for each of them.
We generally agree with the assessment of the potential evolution of automated advice.

Notwithstanding this, it is important that the entities providing the automated advice, particularly the new providers, are properly qualified in order to avoid a negative impact on the provision of the financial advice provided.
The automation will become increasingly part of the provision of financial advice across all the financial sectors. However, this development is still in its early stages and will depend if financial institutions and customers believe it brings benefits for both of them.
We do not have any other comments on this topic.
Elisa Ricón
I