Response to joint Comitteee Discussion Paper on automation in financial advice
Go back
The Joint Committee of the European Supervisory Authorities has recently issued a discussion paper on “automation in financial advice”. This is quite appropriate, since what is also known as “robot advice” is an emerging trend, first appearing in the US about 5 years ago and now slowly coming to Europe.
Better Finance – the European Federation of Investors and Financial Services Users – took a closer look at a sample of these new providers, both in the US and in selected EU countries (Belgium, France, Germany, UK): see annex I for the list of providers researched.
The research concentrated on the products on offer as well as prices (see Table in Annex II). The data collected can be examined in the attached tables. This limited research did not, and could not, analyse the algorithms used by these providers, nor the adequacy of the asset allocations designed and advised to clients. It is of course too early to evaluate the adequacy of the services provided, as investments are still very recent.
The following are the key findings of the research.
1. “Robot advice” in the field of investments: a misnomer
A European legal definition of investment advice can be found in the MiFID Directive: (Article(4)(1)(4)) “'Investment advice' means the provision of personal recommendations to a client, either upon its request or at the initiative of the investment firm, in respect of one or more transactions relating to financial instruments.”
All the providers we researched are indeed providing investment advice under this definition, although the degree of personalisation is debatable: all the providers ask the prospects/clients to fill out a questionnaire on their needs, on their tolerance to risk or on their risk appetite, and often on their existing financial situation, assets and debts. But quite a few of the providers researched then direct customers to a rather limited number of predefined investment strategies or portfolios: three only in the case of Vamoo (Germany), five for Money on toast (UK) and ten for Yomoni (France). Can it really be considered to be a “personal” recommendation when the customer is merely placed in a limited number of pre-defined categories? Actually, the UK and the French providers are duly registered as financial advisors, but the German one is not.
But they are also all also providing asset management services, since they typically implement the personal recommendations provided to their clients by executing the investments proposed and accepted.
In fact, several of the providers researched are registered as asset management companies with their National Competent Authority, or partner with an asset management company.
Therefore, terms such as “robot investing” or “robot investment management” would designate this emerging business more appropriately.
2. Simpler, lower and more transparent fees, but still usually asset-based
As shown in the table in annex II, most “robot” providers display a much simplified fee scale, often a single “all-in” fee, or just two fees: “advice” and fund fees. There typically are no other fees such as entry fees, custody fees, transaction fees, performance fees, wrapper fees, etc. which are often to be found in standard “human” financial advice and private banking services. Fund fees are not always disclosed in the presentation of the robot services but their existence on top of the advice fee is usually mentioned.
The generally far simpler and more transparent fees from robot investing are also much lower than fees charged by “human” financial advisors or private bankers:
• between 12 and 99 “basis points” (fund fees included) in the US (between 0 and 89 bps for advice plus 10 to 17 bps typically for underlying fund fees)
• between 69 and 169 basis points (fund and wrapper fees included) in Europe, with much more recent and smaller providers (between 49 and 100 bps for advice plus 19 to 69 bps typically for funds).
This compares very favourably with traditional overall fees charged by financial advisors and private bankers that are typically far above 100 bps when one includes underlying asset management fees. This difference has a very important impact over time on the actual performance of financial advice, especially in an environment of low capital market returns. It can indeed make the difference between protecting the purchasing power (the real value) of people’s savings and destroying it.
The basis for charging fees, however, remains traditional: it is asset-based, not performance based. This still constitutes a limit to a better alignment of interests between providers and clients. There is a notable exception though: the French robot adviser “Marie Quantier” has developed a rather revolutionary and investor-friendly fee model: € 5,90 fixed per month plus 5% of annual gains.
3. A fee-based model in Europe: a way for low cost index funds to enter the European retail market and for ending conflicts of interests in distribution?
One striking common feature shared by nearly all providers researched is the investment tools they use. They almost exclusively use low cost index funds, predominantly ETF ones. This is actually the principal tool to reduce overall costs and fees, not only the automation technology itself. For instance, instead of using typical “active” retail equity funds (with average annual fees around 70 basis points in the US, and 170 basis points to 180 bps in Europe) they use only low cost index equity funds (most often the “ETF” – exchange traded funds – type), which charge from 5 (US) to 40 bps for developed country equities.
This generates huge fee savings especially in Europe due to the much higher fund fees there. One reason for these higher fees in Europe is the still dominant commission-based distribution model in Continental Europe. This is why the low cost index ETF market still only represents about 10% of the retail market in Europe versus 50 % in the US, since ETFs provide distributors with no or only very small commissions. All the robot providers researched in Europe seem to be fee-based instead of commission based. The fee based model has a very important advantage: it massively reduces the issue of conflicts of interests in the retail distribution of financial products. They exclusively use index ETFs and little else. “Robot advice” is one of the few retail distribution channels in Continental Europe to promote and advertise index ETFs to citizens. Better Finance sees this as a real benefit should “robot” investing develop a significant market share.
4. Overall “robot” investing fees still much higher in Europe than in the US
For example, one French offer stands at 1.60% overall annual fee (including asset management and other fees), which is most probably quite a bit cheaper than traditional private bankers or financial advisers, but still three to six times more expensive than the US leader for example.
Of course, size matters to cover fixed costs and most European robot-investing providers are still in start-up mode. But this also reflects other findings from Better Finance and others indicating that investment–related fees are much higher in Europe (for example, fees 2 to 2.5 times higher for equity investment funds on average). In turn, this is linked to the fragmentation of the European markets and to the lack of product standardisation as well as insufficient competition.
Also, in France, overall fees are increased by the use of costly wrappers : another fee/cost layer is added by wrapping the investment funds into unit-linked insurance contracts for tax reasons. This adds 60 bps per annum to the pricing of Yomoni for example. Better Finance research on the real return of long term and pension savings found that French unit-linked contracts have on average been one of the worst performing long-term financial savings products over the last 15 years in Europe, mainly due to the very high multi-layer fees (2.75% per annum on average for life insurance contracts invested in equity funds ). It seems the tax advantage granted to life insurance unit-linked products in France is more than totally captured by providers, to the detriment of retail clients.
5. Supervision by Public Authorities
Nearly all providers researched were duly registered as financial advisors in their respective jurisdictions like traditional, non-automated financial advisors. In addition, as mentioned above, many are logically also registered as asset managers or have a contractual relationship with a registered investment company.
The German provider is an exception. Vamoo claims that they are not providing investment advice or asset management services but rather act as an investment intermediary, an agent placing clients’ money with their product partner on an execution only basis. The law seems to require for them to submit to Bafin (the German financial watchdog) supervision and the trade supervisory centres (which by the way also control restaurants, bars etc) are the entities in charge of their supervision. Up to now, there seems to be no court case to challenge this classification and we are not aware of any issues raised by Bafin or the trade supervisory centres.
Therefore, and contrary to some critics of “robot” advice, we could not really spot a weaker regulatory framework for automated financial services when compared to traditional financial advice, except for in Germany . Better Finance recommends for the European regulators to more precisely define “investment advice”, and more specifically “personal recommendations”.
We could not, however, test the adequacy and performance of the advice itself, i.e. of the algorithms used. First, because most providers lack the sufficient track record in terms of years of operation, second because the advice is usually fully personalised , and third because they – like most standard advisors – typically do not disclose performance yardsticks .
6. Simple investment advice: “robot” advice does not replace the need for more customised advice
All providers researched mostly offer portfolio advice based on standardised questionnaires. For one, two quite different stated objectives – “regular income” and “retirement” delivered the exact same portfolio advice. The investment management is not always fully personalised as indicated above, as some robot advisers seem to allocate all prospects and clients to a pre-determined and rather limited number of standard portfolios. In that case, the value added compared to risk-adjusted balanced funds or to life cycle funds (for retirement needs) seems more or less limited to the selection of the portfolio/fund, and most often regular rebalancing (but life cycle funds typically do that as well).
There is little taking into account of more complex situations (like varying income needs for example), of tax optimisation, asset/liability management (debt issues), or generational issues, etc. In our view it does not fully replace a face-to-face interview relationship and more customised advice that goes beyond the asset allocation and the selection of investment products for quite simple time horizons.
However, these findings have to be balanced by the fact that several robot investing providers offer the possibility to speak to an adviser over the phone or electronically, in particular for the assessment phase.
Anyway, asset allocation and investment product selection are still an important part of the value generated by financial advice, and “human” financial advisors could most probably benefit from expanding their own use of automated tools.
7. Conclusion
Robot investing is still at a very early stage – especially in Europe - but it does carry some real promises for savers and individual investors.
24. Are there any other comments you would like to convey on the topic of automation in financial advice?
Executive SummaryThe Joint Committee of the European Supervisory Authorities has recently issued a discussion paper on “automation in financial advice”. This is quite appropriate, since what is also known as “robot advice” is an emerging trend, first appearing in the US about 5 years ago and now slowly coming to Europe.
Better Finance – the European Federation of Investors and Financial Services Users – took a closer look at a sample of these new providers, both in the US and in selected EU countries (Belgium, France, Germany, UK): see annex I for the list of providers researched.
The research concentrated on the products on offer as well as prices (see Table in Annex II). The data collected can be examined in the attached tables. This limited research did not, and could not, analyse the algorithms used by these providers, nor the adequacy of the asset allocations designed and advised to clients. It is of course too early to evaluate the adequacy of the services provided, as investments are still very recent.
The following are the key findings of the research.
1. “Robot advice” in the field of investments: a misnomer
A European legal definition of investment advice can be found in the MiFID Directive: (Article(4)(1)(4)) “'Investment advice' means the provision of personal recommendations to a client, either upon its request or at the initiative of the investment firm, in respect of one or more transactions relating to financial instruments.”
All the providers we researched are indeed providing investment advice under this definition, although the degree of personalisation is debatable: all the providers ask the prospects/clients to fill out a questionnaire on their needs, on their tolerance to risk or on their risk appetite, and often on their existing financial situation, assets and debts. But quite a few of the providers researched then direct customers to a rather limited number of predefined investment strategies or portfolios: three only in the case of Vamoo (Germany), five for Money on toast (UK) and ten for Yomoni (France). Can it really be considered to be a “personal” recommendation when the customer is merely placed in a limited number of pre-defined categories? Actually, the UK and the French providers are duly registered as financial advisors, but the German one is not.
But they are also all also providing asset management services, since they typically implement the personal recommendations provided to their clients by executing the investments proposed and accepted.
In fact, several of the providers researched are registered as asset management companies with their National Competent Authority, or partner with an asset management company.
Therefore, terms such as “robot investing” or “robot investment management” would designate this emerging business more appropriately.
2. Simpler, lower and more transparent fees, but still usually asset-based
As shown in the table in annex II, most “robot” providers display a much simplified fee scale, often a single “all-in” fee, or just two fees: “advice” and fund fees. There typically are no other fees such as entry fees, custody fees, transaction fees, performance fees, wrapper fees, etc. which are often to be found in standard “human” financial advice and private banking services. Fund fees are not always disclosed in the presentation of the robot services but their existence on top of the advice fee is usually mentioned.
The generally far simpler and more transparent fees from robot investing are also much lower than fees charged by “human” financial advisors or private bankers:
• between 12 and 99 “basis points” (fund fees included) in the US (between 0 and 89 bps for advice plus 10 to 17 bps typically for underlying fund fees)
• between 69 and 169 basis points (fund and wrapper fees included) in Europe, with much more recent and smaller providers (between 49 and 100 bps for advice plus 19 to 69 bps typically for funds).
This compares very favourably with traditional overall fees charged by financial advisors and private bankers that are typically far above 100 bps when one includes underlying asset management fees. This difference has a very important impact over time on the actual performance of financial advice, especially in an environment of low capital market returns. It can indeed make the difference between protecting the purchasing power (the real value) of people’s savings and destroying it.
The basis for charging fees, however, remains traditional: it is asset-based, not performance based. This still constitutes a limit to a better alignment of interests between providers and clients. There is a notable exception though: the French robot adviser “Marie Quantier” has developed a rather revolutionary and investor-friendly fee model: € 5,90 fixed per month plus 5% of annual gains.
3. A fee-based model in Europe: a way for low cost index funds to enter the European retail market and for ending conflicts of interests in distribution?
One striking common feature shared by nearly all providers researched is the investment tools they use. They almost exclusively use low cost index funds, predominantly ETF ones. This is actually the principal tool to reduce overall costs and fees, not only the automation technology itself. For instance, instead of using typical “active” retail equity funds (with average annual fees around 70 basis points in the US, and 170 basis points to 180 bps in Europe) they use only low cost index equity funds (most often the “ETF” – exchange traded funds – type), which charge from 5 (US) to 40 bps for developed country equities.
This generates huge fee savings especially in Europe due to the much higher fund fees there. One reason for these higher fees in Europe is the still dominant commission-based distribution model in Continental Europe. This is why the low cost index ETF market still only represents about 10% of the retail market in Europe versus 50 % in the US, since ETFs provide distributors with no or only very small commissions. All the robot providers researched in Europe seem to be fee-based instead of commission based. The fee based model has a very important advantage: it massively reduces the issue of conflicts of interests in the retail distribution of financial products. They exclusively use index ETFs and little else. “Robot advice” is one of the few retail distribution channels in Continental Europe to promote and advertise index ETFs to citizens. Better Finance sees this as a real benefit should “robot” investing develop a significant market share.
4. Overall “robot” investing fees still much higher in Europe than in the US
For example, one French offer stands at 1.60% overall annual fee (including asset management and other fees), which is most probably quite a bit cheaper than traditional private bankers or financial advisers, but still three to six times more expensive than the US leader for example.
Of course, size matters to cover fixed costs and most European robot-investing providers are still in start-up mode. But this also reflects other findings from Better Finance and others indicating that investment–related fees are much higher in Europe (for example, fees 2 to 2.5 times higher for equity investment funds on average). In turn, this is linked to the fragmentation of the European markets and to the lack of product standardisation as well as insufficient competition.
Also, in France, overall fees are increased by the use of costly wrappers : another fee/cost layer is added by wrapping the investment funds into unit-linked insurance contracts for tax reasons. This adds 60 bps per annum to the pricing of Yomoni for example. Better Finance research on the real return of long term and pension savings found that French unit-linked contracts have on average been one of the worst performing long-term financial savings products over the last 15 years in Europe, mainly due to the very high multi-layer fees (2.75% per annum on average for life insurance contracts invested in equity funds ). It seems the tax advantage granted to life insurance unit-linked products in France is more than totally captured by providers, to the detriment of retail clients.
5. Supervision by Public Authorities
Nearly all providers researched were duly registered as financial advisors in their respective jurisdictions like traditional, non-automated financial advisors. In addition, as mentioned above, many are logically also registered as asset managers or have a contractual relationship with a registered investment company.
The German provider is an exception. Vamoo claims that they are not providing investment advice or asset management services but rather act as an investment intermediary, an agent placing clients’ money with their product partner on an execution only basis. The law seems to require for them to submit to Bafin (the German financial watchdog) supervision and the trade supervisory centres (which by the way also control restaurants, bars etc) are the entities in charge of their supervision. Up to now, there seems to be no court case to challenge this classification and we are not aware of any issues raised by Bafin or the trade supervisory centres.
Therefore, and contrary to some critics of “robot” advice, we could not really spot a weaker regulatory framework for automated financial services when compared to traditional financial advice, except for in Germany . Better Finance recommends for the European regulators to more precisely define “investment advice”, and more specifically “personal recommendations”.
We could not, however, test the adequacy and performance of the advice itself, i.e. of the algorithms used. First, because most providers lack the sufficient track record in terms of years of operation, second because the advice is usually fully personalised , and third because they – like most standard advisors – typically do not disclose performance yardsticks .
6. Simple investment advice: “robot” advice does not replace the need for more customised advice
All providers researched mostly offer portfolio advice based on standardised questionnaires. For one, two quite different stated objectives – “regular income” and “retirement” delivered the exact same portfolio advice. The investment management is not always fully personalised as indicated above, as some robot advisers seem to allocate all prospects and clients to a pre-determined and rather limited number of standard portfolios. In that case, the value added compared to risk-adjusted balanced funds or to life cycle funds (for retirement needs) seems more or less limited to the selection of the portfolio/fund, and most often regular rebalancing (but life cycle funds typically do that as well).
There is little taking into account of more complex situations (like varying income needs for example), of tax optimisation, asset/liability management (debt issues), or generational issues, etc. In our view it does not fully replace a face-to-face interview relationship and more customised advice that goes beyond the asset allocation and the selection of investment products for quite simple time horizons.
However, these findings have to be balanced by the fact that several robot investing providers offer the possibility to speak to an adviser over the phone or electronically, in particular for the assessment phase.
Anyway, asset allocation and investment product selection are still an important part of the value generated by financial advice, and “human” financial advisors could most probably benefit from expanding their own use of automated tools.
7. Conclusion
Robot investing is still at a very early stage – especially in Europe - but it does carry some real promises for savers and individual investors.