Yes, but risk should also be discussed in the context of return. Too much focus on “possibility of capital loss” and “uncertainty of return” could lead customers to avoid all risks. There is also a risk in avoiding all risk, a zero risk investment normally has a negative real return in the long run.

Do you agree that market, credit and liquidity risk are the main risks for PRIIPs?
Yes
Do you agree with the definitions the ESA’s propose for these?
Yes

Risk measurement is, in our view, best quantified by historical data. It could be calculated on the existing product, if available, or by the historical data of the investment type/underlying (for example, equity, bonds, MM and so on. History may of course not be a perfect predictor of future risk but this problem of uncertainty of the future is not solved by forecast models. Instead we believe that the shortcomings on using history in the risk measurement could be offset by a narrative explanation.

Although probabilistic modelling could help the customer in assessing likelihood, we dismiss it since it gives the customer a false sense of security. It is better to use a (headline) text like “Let´s say, hypothetically, that” show how the product would perform under one or more scenarios. Scenarios should be in line with/have some kind of relevance to (if any) historical data.
The parameters assumed should be coordinated with the cost section.

Modelling would have to involve too many complex and uncertain parameters.
Due to the difference in pricing structures all scenarios should be presented net of costs.

We have no specific view on time frames, although for comparison reasons they should be expressed in “percentage per year” in order to be able to compare products with different holding periods. Percentage per year is also a concept most consumers are familiar with, think of interest on loans and return on deposits. We would prefer some sort of connection to the products recommended (minimum) holding period.
The parameters assumed should be coordinated with the cost section.

We prefer a combination: aggregate monetary amounts and annualized percentages. This is also consistent with the cost section.

Arguments claiming that net of costs is practically hard to present normally origins from complex fee structures, constructed in order to make it hard for the consumer to see the net return. We strongly prefer net of costs. A (gross) scenario that needs a “prominent warning” and needs to be cross read with the cost section is of little help to the investor. Performance is by definition net of costs, especially from a consumer viewpoint that this KID is supposed to deliver.

We prefer the UCITS KII risk and reward indicator, mainly since it captures both perspectives: risk and reward. The Dutch indicator in the financial leaflet is in our view especially harmful, since it gives the impression that the best is to take on as little risk as possible, which is not always in the best interest of the consumer.

We believe that past performance should be the base for performance scenarios. If no or too short history, methods should be standardized (for example return for the market as a whole, return for the product type as a whole).

We prefer 2B or 3B, a single visual element for performance and single or multiple visual elements for risk. The A-alternatives will lead to too much focus on risk and the C-alternatives will be too complex.

Fairly ok but it should be kept in mind that it is too ambitious to believe that a 3 page KID could give answers to 8 “key questions” and 21 follow-up questions on costs. Prioritization is needed, preferably on key question 1 and its follow up questions.

It is fairly easy to list all costs. The main challenge is to find a standardized way to present them in order to make a comparison among products. Standardized rules for presentation are the best way to address this. The rules must be not benefit any particular fee structure.

Intrinsic value seems to be the best approach, but it needs a standardized mechanism for estimating fair value.

We prefer (an improved version) of the “ongoing charges”.
The RIY calculation is interesting but has some methodological problems. For example it includes alternative costs (which is not a cost paid from the investor to the provider) that is highly sensitive to the performance assumptions. For example, a scenario with negative performance will end up with a “negative cost” (i.e. an income) for the investor if the alternative cost is included.

We think that the growth rate should be consistent with the performance scenarios. A growth rate higher than 0 (which is likely in the long run) will give both a higher performance and (in the case of most cost models) a higher cost. Both need to be shown in order to give the investor a balanced view.

Regarding the summary indicator: We prefer Option 2, because it is simplistic and easy to understand and gives the investor a sense of what is to be considered price worthy vs. expensive.
Regarding the aggregated cost figures: We prefer option 6, but question whether the “What might you get back at 5%” column is needed. In any case, it needs to be stressed that it is an “assumed example” so that the investor does not believe it is a promise.
(But we do not understand the figures in the example. For example, in year 1, if you have invested £5000 and had a 5% yield and £328 in costs, you would have £4922 back)
Most of the other options are far too complicated.

On the one hand aggregation is better since it will be easier to communicate. On the other hand, no aggregation is better since we believe that the risks are so different in kind that they need to be kept separate in order to be understandable. We have a slight preference for a multidimensional summary risk indicator, but consumer testing is the way to decide this.

We basically agree but have a reservation on taxes. You could argue whether taxes should be considered as costs. But if they are considered as costs, why only transaction tax and stamp duty and not other taxes (such as tax d’abonnement and VAT)?

SIFA would recommend that the same requirements on identity information as in the Commission Regulation (EU) No 583/2010 (the KIID regulation) article 4.4 – 4.6 are applied on the products covered by the PRIIPs regulation.
Contact details: Whereas in the KIID regulation the contact details of the KIID may be included in the section “practical information” (see article 20), contact details are requested in the beginning of the PRIIPs document. SIFA would recommend that such information is standardized and kept short.
Information about the competent authority: Regarding information about the competent authority SIFA suggests that only the name and a current web-address should be requested.
Universal identifiers (ISIN): SIFA is of the opinion that ISIN or other identifier, should be included where available .
Other: SIFA is also of the opinion that the document should contain the date of its publication.

SIFA is of the opinion that the ESAs should clarify what instruments do not require a comprehension alert. The ESAs should design a non exhaustive list defining those products where no comprehension alert is necessary. Products that are generally considered consumer products should not require a comprehension alert to avoid confusion among investors.

SIFA is of the opinion that legal form is the most appropriate way to classify products.

SIFA agrees that the descriptions should be kept brief and if prescribed statements are needed there should still be room left for specific descriptions appropriate for certain fund types.

SIFA would like to highlight the possibility of describing the process of investing used by the product manufacturer. Language should be kept simple.

SIFA agrees with the ESAs discussion under this section. SIFA would especially like to stress what is acknowledged about identified target market within MiFID II. Consistency with MiFID II in this area is very important to avoid confusion among investors.

Under this section investor compensation schemes will be explained. Such schemes only present one possible way of protecting investors. Other ways such as those where investors are protected through separation of investments and manager or manufacturer must be explained and compared with the protection of other products. It should be highlighted that funds for example have depositaries that provide this oversight function that no other products have.

Exact generic timeframes are difficult to produce since the timeframe would differ depending on product and investor.

It is important to stress that the product manufacturer can only provide information about how to complain about the specific product.

SIFA agree with the ESAs that the UCITS KIID should be taken as a starting point for questions on periodic review, revision and republication.

Regular updates of the PRIIPs KID should be made public through broadcasting on a web-page.

SIFA agrees that it is important that investors are given the opportunity to consider the KIID. However, it has to be balanced against the investors need to make his or her own decision about how much information is needed. Overly prescriptive regulation does not acknowledge each individuals right to make his or her own decision about the timing of an investment. The KIID requirements to provide the document in good time before the investment should be sufficient.

SIFA is of the opinion that the UCITS requirements should also be taken into account when assessing the timing of the delivery of the KIID.

SIFA is in favour of templates and has developed templates for its members to aid in the development of UICTS KIIDs. However, such templates should be examples, making it possible for manufacturers to use them as ‘safe harbours’ but to deviate if that is more appropriate. Several templates should be provided.