We agree that market, credit and liquidity risk are the main risks for PRIIPS and agree with the definitions proposed.

The vzbv strongly doubts that any models predicting future performances are in any way reliable. In this case it does make no difference if modelled or historical data is used. We understand that customers seek confirmation and want a clear performance forecast but this is impossible to deliver, if you prefer reliable results. Even the providing of historical data without a clear prediction can be misleading, because investors will often falsely assume, that this data may be relevant for future performance. This phenomenon is described by behavioral economists under the terms ‘framing’ (depending on the presentation and order of presentation of information), ‘availability’ (depending on the access to old/early information and data) and ‘representativeness’ (depending on previous experience).
Because future performances are only hard to predict, we prefer a narrative presentation of the expected yield and factors and parameters on which it depends on. This includes:
• Developments and other assets on which the performance depends on.
• Potential costs which decrease the overall yield.
• Asset categories in which the capital gets invested in.
There have already been several scientific studies in an experimental environment, which support our arguments. To verify this position we propose that there will be further consumer testing in a real market environment, executed by the ESAs.

We support an approach using several holding periods for the performance scenarios. They must accommodate a clear distinction between performance scenarios in the short and long run.

The performance scenarios should be foremost presented in monetary amounts, because retail investors can relate to them the best. Additional relative figures can be only complimentary to the monetary figures.

We would like to stress here the importance of having costs accounted for in the performance scenarios, as this is essential for the retail investor to understand the real amount he could be receiving in the end. Any reference to “intrinsic (or gross) performance” should only be added additionally to the expected real payout for the retail investor.

We feel that disclosing three presented scenarios with projected probabilities would serve the necessary variety in scenarios, while not confusing retail investor with abundant information.

As laid out in our answer to Q5, a good overall risk indicator is of utmost importance. The Belgian example serves here as a very good example. Such a presentation does a better job in alerting consumers to potential risks than the classic numerical approach of the UCITS KIID.
Furthermore additional options for separately displaying credit or liquidity risk need to be assessed.

For the sake of clarity, we prefer an easy to understand single visual element which presents a plausible best, plausible middle and plausible worst case. The example on page 43 (thermometer tested in the Netherlands) is a good example.

We basically agree with the description of the consumer’s perspective on costs as expressed in the key questions. However, one helpful additional dimension would be to disclose the nature of the costs expressed in the KID. For consumers, it is important to understand why some costs exist and consequently evaluate if a particular cost is of administrative nature or whether it reflects a substantial fee/commission/margin taken by the issuer/distributor of the product.

The existence of different (national) markets and the related disparities in the types of PRIIPS marketed, pose a challenge towards ensuring a level-playing field in cost disclosures on a European level. But in general, we believe that the adoption of an overall cost indicator, followed by a detailed cost breakdown, is the good way forward. The overall aim should be giving the retail investor already an ex ante, full picture on how much he will have to pay for a certain product.
In order to ensure comparability, cost disclosures based on standardized investment amounts deserves further consumer testing.
As for structured products, the inclusion of implicit costs is another key challenge. Measures like the presented “intrinsic value” and “issuer estimated values”, which also include the product issuer’s margin can address the overall cost disclosure and are easily understood. The additional disclosure of single types of costs would allow to compare the costs in detail.

We would like to stress first that by all means implicit costs should be taken into account, in order to close loopholes for product manufacturers, rendering any reliable comparison between products very difficult.
Implicit costs can be accounted for by using the above mentioned measures of “intrinsic value” and “issuer estimated values”.

In our eyes the approaches using fair values are well suited methods for preparing “total aggregate cost” figures but its use seems somewhat limited to certain implicit costs. The total expenses ration (TER) method does not include entry or exit costs, performance fees, or portfolio transaction costs. Therefore we think that it can be misleading for customers and do not evaluate it fitting to aggregate costs in KIDs.
One of the problems with the reduction in yield (RIY) approach is, that it does not account for biometrical costs and the value of the RIY depends on the time frame of the PRIIP and its return. This makes it very hard to compare different PRIIPs and can lead to confusing results. Furthermore the RIY is displayed in relative values which shrouds the real value for the investor and it does not include guaranteed costs. Another problem with RIY is that the costs it includes are dynamic and not static. They change during the investment (e.g. issue surcharges or success related costs) and have to be estimated.

The reduction in payment (RIP) approach addresses the issues with other approaches mentioned in question 20 and can also be (and should be) expressed in monetary values via an easy calculation. Therefore we think it is the most suited method to prepare “total aggregate cost” figures.

The Norman key figure is a good an easily understandable tool for comparing the impact of costs.

Market, credit and liquidity risk each have a different impact, depending on the type of product and the type of retail investor involved. The liquidity risk for a retail investor with a long investment horizon taking out a 10 year fixed maturity product is not so important. But the absence of any secondary market for the same product could be very relevant for a retail investor with higher liquidity needs.
At the same time, many consumers prefer having a single indicator which sums up the overall risk.
Therefore we would advocate for having at least one good visual indicator summarizing the overall risk, together with a separate mentioning of liquidity and credit risk (either by a narrative explanation or by a separate visual).

We would like to add one type of cost to table 12: the issuer’s margin/mark-up should be included as an important part of the costs faced by the consumer. This would also be in line with the pending MIFID II requirements.
In practice, different issuers can use different margins for identical products, which is very hard to discern for a retail investor with few experience. Therefore we insist on treating the margin as a separate cost so the consumers can take this into account as well.

We have some reservations about using the list, deriving from art 50 of UCITS, as a proxy for assets in which retail investors commonly invest. We would like to emphasize that UCITS are highly diversified products which invest in a broad range of asset classes. Accordingly, one cannot derive from this proposition that retail investors necessarily have experience in investing directly in particular assets from the UCITS horizon.
We would therefore advise that recital 18 should be interpreted as following:
• It invests in underlying assets that are not commonly invested in directly by retail investors.
Finally we would like to point out that the comprehension alert should be placed very near to the summary risk indicator of the KID.

We do agree with the assessment of when PRIIPs might be concerned by article 6(3) and we agree with the expectation of the Joint Committee that these products are mainly unit-linked life insurance contracts and hybrid life-insurance-contracts.

When multiple KIDs are used to inform customers about a product with many options it is a high priority for us, that the presented options become as comparable as possible. Examples can be a good way to do this and they can help to understand the product easier. But these examples need to be designed in a coherent way for the different options to ensure comparability.

Instead of the use of ranges for the presentation, the correspondent indicators for the different options could be displayed within one graphic respectively for risk, performance and costs.

In general we feel that the ESAs should develop a fixed template, as it is a key ambition of the KID is to foster comparability between investment products. Some flexibility to cater for the different product classes can be considered, but this has to be kept minimal in order not to provide loopholes for hiding costs.