EIOPA IRSG (Insurance and Reinsurance Stakeholders Group)
Maria Aránzazu del Valle
It is understood that these Key questions have been identified only to describe the consumer perspective rather than for the purpose of being disclosed as such in the KID. The only question related to the risks that should appear in the KID is “what are the risks and what could I get in return?” as established by Article (8)(3)(d) of the PRIIPs Regulation. Thus, it is important that the questions correspond to and do not go beyond the provisions of the PRIIPs Regulation.
It must be highlighted that some of the Key Questions use informal and negative wordings and are expressed in the first person wrongly creating the impression that the KID contains personalised information rather than general pre-contractual product information. If they are to appear in the KID, some of them should be reworded to reflect a more neutral approach. This is particularly the case when compared to the requirements for UCITS funds. It is of utmost importance that risks and reward features of PRIIPs are properly balanced.
As an example, informal questions such as “how much can I win?” should be reformulated to ensure that it is understood that many PRIIPs products (especially most of insurance-based investment products) are investment products as opposed to pure speculative products.
Another example could be the question “is risk and return balanced?” that encompass a judgement element which should be left to consumers (this will depend of the risk profile of each investor).
Regarding the question “What are the risks and what could I get in return”, and from the consumer perspective, it seems reasonable that the information on risks and expected return (performance) is reflected separately, as the combination of both in the same chapter/line may result in the consumer paying more attention to the information on expected return rather than to the information on risks. It should be stressed in the KID that the return is expected. However, it should be made clear that there is a trade-off between risk and reward. The situation in which the consumer only pays attention to the information on risks (disregarding the information on expected return) should also be avoided.
With regard to the section “How long should I hold it and can I take my money out early?”, it is also essential that the information on the consequences of cashing in early (for example: additional costs or penalties, the possibility of losing the capital protection, additional contingent fees) is clear and not misleading for the customer.
The proposed definitions could be a good starting point. However, not all these risks are relevant for each investor. It will depend on the product itself:
For most of insurance-based investment products market risk is the most relevant factor. The impact of inflation on the value of the PRIIP should not be included as one aspect of the market risk mainly because inflation is not a risk that is inherent only for PRIIPs but affects other investment products that are excluded from the scope of PRIIPs (real estate, simple bank deposits, equities, fixed income) in the same way (level-playing field issue). When necessary a general short narrative explanation of the effects of inflation could be included.
As far as credit risk is concerned, the Solvency II regime ensures the financial soundness of insurers. Therefore credit risk is less relevant for insurance-based investment products than for other PRIIPs (derivatives, structured deposits or securities). Furthermore, insolvency guarantee schemes should be taken into account when assessing the credit risk.
Long-term insurance-based investment products are being purchased notably because of the very characteristic that they are long-term, which is a feature of the product rather than a risk. For such a product, the liquidity risk should not be assessed in the same way as other short term investments.
Relying on quantitative measures alone could be misleading for customers. If applied, they should always be complemented by qualitative measures and generic criteria. In any case, the use of quantitative measures, qualitative measures or a combination of both will depend on the product itself. A main conclusion that can be drawn from this is that it is completely necessary that different KID templates are used for the different types of PRIIPs to the extent this does not impair the general comparability among them.
Regarding insurance-based investment products, it should be taken into account that at inception the PRIIPs regulation was focused only in insurance products where the invertor bears the investment risk (Unit-Linked/Index-Linked products, that are substitute products of UCITS), for which the KID can be more easily adapted.
The subsequent inclusion of “traditional” life insurance products (more similar to simple deposits or fixed income bonds that are excluded from the scope) requires a greater adaptation of the KID to reflect the specific nature, risks and characteristics of this type of PRIIPs.
Regarding the possible quantitative measures of market risk, while measures like historical volatility, VaR, ELVaR, etc., could be appropriate for some types of PRIIPs (those that are similar to UCITS), it could not be appropriate at all for other types of PRIIPs.
For life insurance products that guarantee a fixed interest rate at maturity, with fixed income bonds as underlying assets, a classification of market risk according to the underlying investments could be much more reasonable.
Notwithstanding the above, it should also be highlighted that additional guarantee mechanisms can change the risk of the product significantly compared to that of the underlying assets, so that just measuring the risks of the underlying assets could not be appropriate.
The same occurs in the case of possible measures of the credit risk. Clearly quantitative measures like credit spread or CDS spread of the manufacturer, credit value at risk or qualitative measures like credit rating are not applicable for most of insurance-based investment products. Many life insurance companies are not quoted or listed or don´t have a credit rating issued by a credit rating agency. It must be reminded that there are several EU initiatives in progress to reduce the overreliance on credit ratings. Past experiences (Lehman Brothers was rated AA just before collapsing) demonstrate that credit rating is not always a reliable measure of credit risk. Prudential supervision should be the more relevant qualitative measure for insurance-based investment products. We agree that a clear distinction should be made between entities subject to prudential supervision (credit entities, insurance undertakings) and other entities.
With regard to possible measures of liquidity risk, it is clear that quantitative measures like the bid-offer spread or the average volume traded are meaningless for insurance-based investment products, for which only qualitative measures explaining the consequences of cashing in early are possible. For Unit-Linked products in which the investor bears the investment risk attention should be paid to the liquidity of the underlying assets. In any case, for products which are explicitly designed and marketed for a longer time horizon (e.g. old-age provision products), the liquidity aspects should not be over-emphasized.
These should be addressed as costs only paid in specific situations or under specific conditions, i.e. they should be presented separately.
Practice and consumer testing have shown that probabilistic modelling is often not understood by consumers as opposed to deterministic modelling. Therefore, a deterministic modelling seems much more suitable for the performance scenarios. When an uneven number of scenarios are shown, it should be made clear that the middle scenario could not be the most probable.
Realistic scenarios are necessary to reflect the usual long-term nature of this type of PRIIPs. Moreover, since risk, reward, performance and costs of an insurance-based investment product are strongly correlated, a consistent approach and presentation of these features are needed.
It should also be taken into account that the KID does not contain personalised information but general pre-contractual product information for an “average” customer.
Regarding the KID template for insurance-based investment products, it should be taken into account the usual long-term nature of this type of PRIIPs. The time frame of the scenarios should be consistent with the maturity/duration/timeline of the product or, if unknown, with the recommended holding period. The consequences of early surrender should be discussed in another section of the KID (How long should I hold it and can I take money out early?).
When performance scenarios include absolute figures or monetary amounts to increase comprehensibility it should be made clear in the KID that the consumer is not receiving personalized information but only general pre-contractual product information for an “average” customer.
It is of utmost importance that the performance scenarios are consistent with the information on costs included in the cost section of the KID so that none of the features of the PRIIP is accounted for twice. A fully consistent approach and presentation of performance scenarios and costs is essential.
It seems reasonable that the performance scenarios (average annual returns expressed in percentages) are presented net of costs, but this probably means that certain integration between the sections “performance scenarios” and “costs” is needed. This objective could be achieved with a Reduction in Yield (RIY) approach. RIY is a method for expressing the overall impact of costs in terms of their negative impact on a notional gross yield for a product.
Only certain costs (not contingent costs) should be considered for the performance scenarios. Cost profit participation of the customer should be taken into account, such that the customer gets a realistic scenario.
Regarding insurance-based investment products, the premium for additional insurance benefits (insurance protection against death, disability, etc.) should not be regarded as a cost.
It could be misleading for retail investors if the same number of scenarios was required for all products. The correlation of risk and reward is essential: the number of scenarios should depend on the type of PRIIP.
While the methodology behind the UCITS KII synthetic risk and reward indicator -SRRI-(historical ex post volatility) could be appropriate for some life insurance products, it could not be appropriate for other life insurance products. For many life insurance products the presentation of the UCITS KII SRRI (a scale from 1 to 7) could be supported for the sake of comparability but not the methodology behind it.
For most of insurance-based investment products, the credit and liquidity risks should not be included in the summary risk indicator, which would only reflect the main relevant risk factor (market risk).
From the consumer perspective, a visual risk indicator (using figures, letters, graphs or colours) could have some advantages as it attracts consumers´ attention and may be easier to understand.
Notwithstanding the above, it should also be taken into account that the use of a summary risk indicator that classifies the risk level by colours could imply a significant cost for manufacturers because they will have to buy a colour printer for every distribution point. Therefore, the risk classification by colours should be avoided (the use of numbers or gray scales could have the same effect for consumer but the adaptation costs for manufacturers could be much lower).
Additionally, it should be made clear to customers that there is always a trade-off between risk and reward. The product should match the customer type and his/her risk profile. The use of the red colour could imply that a product with high risk is always negative, whereas this is not necessarily true for some customer types willing to take higher risks in exchange of the possibility of achieving a higher return (or a higher loss). Risk/reward indicators should not be confused with energy labels where high energy consumption is always negative.
Answers to questions 6 to 11 show the difficulties for an abstract presentation of performance scenarios. It is much more relevant a fully consistent approach and an integrated presentation of performance scenarios and costs.
It doesn´t seem very realistic the combination in a single or multiple visual element (abstract presentation) of risks and performance scenarios. However, the outcome of the consumer testing will be important in order to assess the benefits and disadvantages of the different possible options.
Similarly to the key questions related to consumers’ perspective on risk, some the key questions on cost identified in the discussion paper in table 10 should not be disclosed as such in the KID. The only question related to the costs that should appear in the KID is “what are the costs?” as established by Article (8)(3)(f) the PRIIPs Regulation.
Surprisingly, a very relevant key question for the customer (What is my guaranteed/expected annual return net of costs?) is not reflected in Table 10.
It is of utmost importance that the information on costs included in the cost section of the KID is consistent with the performance scenarios so that none of the features of the PRIIP is accounted for twice. The integration of performance scenarios and costs is essential (much more that the integration of risk indicators and performance scenarios).
The premium for additional insurance benefits (insurance protection against death, disability, etc.) should not be regarded as a cost.
The cost structures of each type of PRIIP (derivatives, structured deposits, UCITS, life insurance contracts, etc.) are very different. Therefore, this is the section that will likely require the greatest adaptations in the KID template of each type of PRIIP. Within investment-based insurance products, cost structures could also be very different.
The objective of achieving a level-playing field should be balanced with another relevant objective which is reflecting adequately the cost structures of each type of PRIIP.
Insurance PRIIPs are not traded. Therefore market valuation is difficult / impossible. Additionally, for many insurance PRIIPs there aren´t always dedicated assets for each contract.
A RIY (expressed in percentage) could be an appropriate method to disclose the total annual aggregated cost (costs per year) for one or more scenarios:
Scenario 1 Scenario 2
Gross average annual return: 3% 1.5%
Costs per year: -0.75% -0.5%
Net average annual return: 2.25% 1.0%
As mentioned before, it is essetial that the information on costs included in the cost section of the KID is consistent with the performance scenarios so that none of the features of the PRIIP is accounted for twice. A consistent approach and presentation of performance scenarios and costs is essential (see example in question 20)
The main challenge would be to find a similar format for products that have totally different costs specificities. It should be acknowledged that often these products are not substitutes and therefore the format should not necessarily be exactly similar if it is in the interest of consumers. Establishing a methodology behind the cost disclosure will also represent an important challenge.
See answer to question 16.
Option 5 (page 68) could be a good starting point should it present the total costs per year expressed in percentages rather than for the whole investment period.
Presenting the total costs per year also in absolute figures/monetary amounts should be avoided as they could potentially give the impression that the consumer is receiving personalized information rather that general pre-contractual product information for an “average” customer.
As not all the PRIIPs will have the same maturity/holding period (1 year vs. 20 years), the presentation of average annualized costs is the only way to guarantee comparability.
With the introduction of the above-mentioned amendments, Option 5 seems simple and precise as well as taking into account the PRIIPs Regulation (ie. Possibility of split between entry, ongoing and exit costs). Option 6 could also be a good starting point.
It doesn´t seem very realistic that market, credit and liquidity risk can be integrated in a summary risk indicator. As mentioned before, not all risks are relevant for each type of PRIIP.
For most of insurance-based investment products only the market risk could be captured by a summary risk indicator through qualitative measures, complemented with narrative explanation. Qualitative information regarding the credit and liquidity risks could be added within the narrative explanation of the risks if they are materially relevant for the product. In any case, early surrender should be discussed in another section of the KID (How long should I hold it and can I take money out early?).
Since most of insurance-based investment products are long-term products, only average annualised costs make sense. This becomes particularly obvious if products that have a term of 1 year are compared with products that have duration of 30 years. Therefore, the representation of annualised costs together with a reduction in yield approach could be the most appropriate method for the cost representation, which is also very useful and understandable for the consumers.
The premium for additional insurance benefits (insurance protection against death, disability, etc.) should not be regarded as a cost.
A level playing field must be sought for PRIIPs and UCITS. The transaction fees need not therefore be disclosed until also disclosed for UCITS funds.
Further details as to which products should apply this comprehension alert would be necessary. In order to ensure that the comprehension alert is helpful from the consumer’s perspective, it should apply only to the most complex PRIIPs products.
The comprehension alert would indeed lose its value and would not help consumers if it is used for a wide range of products including some that should not fall under its scope.
Regarding the insurance-based investment products, the comprehension alert should not be applicable for most of them, with the exception of Unit-Linked products o similar products in which the investor bears the investment risk with very complex underlying assets.
The KID should allow the customer to distinguish the legal type of PRIIP (derivatives, structured deposits, UCITS, life insurance contracts, etc.).
Descriptions should be kept very brief. It is important to keep in mind that the document must remain concise (limited to 3 pages) to ensure that the information remains helpful for consumers.
Information could include a generic description of the consumer type at whom the PRIIP is aimed at, but it should not be regarded as a personal recommendation. Selling practices are not part of the PRIIPs regulation that will be established, as regards to insurance-based investment products in IMD 2.
Insurance-based investment products have per se an insurance component. Consumers should be informed that, in addition to offering an investment opportunity, an insurance PRIIP (unlike other PRIIPs) provides for additional protection against death, disability, grave illness, etc.
The premium for additional insurance benefits (insurance protection against death, disability, etc.) should not be regarded as a cost for the performance scenarios.
Regarding insurance-based investment products, there are countries in which this kind of products represent the majority of the market and there are other countries in which they represent a minority compared to traditional life insurance products.
Regarding insurance-based investment products, the products concerned are mainly Unit-Linked life insurance contracts.
It is completely necessary that different KID templates are used for different types of PRIIPs to the extent this does not impeach the comparability among them. A classification according to the legal form of the contract or instrument seems to be appropriate. It is important that the KID only includes information which is relevant for the specific product. Therefore, the information should be tailored to the specific features of insurance-based product. Even a finer distinction between different investment-based insurance products (Unit-Linked / traditional life insurance products) could be necessary.
We agree that annuities may be a special case and that they may be exempt by virtue of being retirement vehicles.
Regarding the use of a summary risk indicator that classifies the risk level by colours, it should be taken into account the significant cost of buying a colour printer in every distribution point. Therefore, the risk classification by colours should be avoided.