Final draft RTS on PRIIPs KID presentation, content, review and provision

JC 2016 21
31 March 2016
Final draft regulatory technical standards with regard to presentation, content, review and provision of the key information document, including the methodologies underpinning the risk, reward and costs information in accordance with Regulation (EU) No 1286/2014 of the European Parliament and of the Council

2
Contents

Executive Summary 3
Background and Rationale 6
Draft RTS 10
Accompanying Documents 78
4.1 Impact Assessments 78
4.2 Feedback on public consultation 140

Executive Summary
These draft Regulatory Technical Standards (RTS) have been developed by the European Supervisory Authorities (ESAs: EBA, EIOPA and ESMA) pursuant to the Regulation (EU) No 1286/2014 on Key Information Documents for Packaged Retail and Insurance-based Investment Products (hereafter, PRIIPs Regulation). The draft RTS relate to three Articles under the PRIIPs Regulation:
 the presentation and content of the Key Information Document (KID), including methodologies for the calculation and presentation of risks, rewards and costs within the document, under Article 8 (5);  the review, revision, and republication of KIDs, under Article 10 (2); and,  the conditions for fulfilling the requirement to provide the KID in good time, under Article 13 (5). The draft RTS text and accompanying Annexes set out proposal in these areas. They reflect the responses to a final Consultation Paper (JC/CP/2015/073) published on the 11 November 2015, which itself followed a general Discussion Paper (JC/DP/2014/02) published on 17 November 2014, and a Technical Discussion Paper (JC/DP/2015/01) published on 23 June 2015. The draft RTS have been developed in view also of the results from a Consumer Testing study conducted by the European Commission. 1 In view of the technical difficulties in developing comparable and comprehensive risk, performance and cost disclosures, the ESAs have also drawn on technical input from a Consultative Expert Group. 2 The draft RTS combine the measures related to Article 8 (5), Article 10 (2) and Article 13 (5). The draft RTS under Article 8 (5) specify the presentation and content of the KID and apply on the PRIIP manufacturer when drawing up a KID. They address the different sections of the KID set out in Article 8 (3), and the underlying methodologies necessary for obtaining and calculating information to be included the KID, for instance in relation to risks, rewards and costs.  It includes a mandatory template to be used, including certain mandatory texts. The template includes details of the layout that must be followed.  For the risk and reward section of the KID, the draft RTS requires a summary risk indicator (SRI) that comprises seven classes, the format of which is to follow the template contained in the RTS.

1 http://ec.europa.eu/finance/finservices-retail/investment_products/index_en.htm. 2 https://eiopa.europa.eu/Publications/Consultations/Consultative Expert Group on Key Information Document for PRIIPs.pdf

4
 The draft RTS also contains a methodology for the assignment of each PRIIP to one of the seven classes in the SRI, and for the inclusion of narrative explanations, and for certain PRIIPs, additional warnings.  The draft RTS includes requirements on performance scenarios and a format that must be followed for the presentation of these scenarios. These are to be presented in tables, showing possible performance for different time periods and at least three scenarios. There are also requirements on the calculation of the figures to be included.  The draft RTS also cover requirements on the presentation of costs, including the figures that must be calculated and the format to be used for these. Detailed methods are included for the measurements and calculations needed for completing the prescribed format for each PRIIP. The cost figures include a presentation of the accumulation of the costs in monetary and percentage terms for standardised period(s), and a breakdown of these costs in percentage terms. The measures for the SRI, performance scenarios and cost disclosures have been developed in view of promoting effective comparability for retail investors between different PRIIPs, and ensuring the information provided to retail investors is as comprehensible for these investors as possible. For these reasons, the presentation of the information has been developed to be as simple and clear as possible, reflecting results from consumer testing. The required simplicity of presentation, notably in relation to the SRI, whilst ensuring sufficient comparability and objectivity between different PRIIPs, has necessitated the development of detailed methodologies to be followed by the PRIIP manufacturer. The ESAs are conscious that it will help PRIIP manufacturers and other stakeholders if supporting material is published on the details of the methodologies, for instance in the form of ‘FAQs’ or ‘How-To Guides’. The ESAs are now starting work on developing supporting material, to ensure as easy an implementation of the measures as is possible. For PRIIPs that offer multiple investment options to the retail investor, the draft RTS set out two approaches that the PRIIP manufacturers may follow:
 Separate KIDs are produced for each option, containing information about both the PRIIP in general and about the option in particular;  The information that would normally be in a single stand-alone KID is separated. A generic KID is produced for the PRIIP in general, and specific information is provided in a separate document or documents about the details of the options (including on their investment objectives, their risks and rewards, and their specific costs). The draft RTS for Article 10 sets out requirements on the PRIIP manufacturer for the revision and republication of the KID at least each year, and an obligation to conduct ad hoc revisions, when necessary under the detailed methodologies for calculating the summary risk indicator, the performance scenarios and the costs, and when necessary for products offering multiple options. The information necessary for PRIIPs traded on exchanges has been considered in view of obligations related to the update and revision of the KID. In this regard, the information required

5 in the KID has been calibrated so as to avoid a situation in which continuous (‘real time’) updating is needed. The KID is conceived as a pre-contractual document, rather than a ‘contract note’ or similar statement of the specific details of a proposed transaction. In this view the information within the KID has been carefully developed to ensure the KID can have general applicability and does not need to be updated on a continuous basis. This includes specifically for exchange traded futures, calls and puts, where the KID requirements have been calibrated to take into account the potential sensitivity of the derivative to conditions on the exchange which otherwise would have required a continuously updated KID. The draft RTS for Article 13 sets out requirements on the person selling or advising on the PRIIP for the KID to be provided sufficiently early for a retail investor to be able to take its contents into account when making an investment decision. The timing of the delivery of the KID can vary depending on the PRIIP in question and the needs of the retail investor. Quantitative and qualitative aspects concerning the costs and benefits of the proposed rules are discussed in section 5.1. This information supplements the proposal and illustrates the reasoning behind the policy choices made. Feedback on the public consultations is given in section 5.2.

6
2. Background and Rationale
The KID for PRIIPs was introduced by the PRIIPs Regulation. The explanatory memorandum to the proposal of the European Commission outlined the intention to improve the comparability and comprehensibility of information being provided to retail investors in the EU, as an important step to improve the protection of retail investors, to aid those investors in comparing between different PRIIPs, and to support the European single market. The PRIIPs Regulation established the main principles to govern the KID, but left a number of important technical details to level two measures, contained in empowerments to the ESAs working together in the Joint Committee. The KID is to be a short – no longer than three sides of A4 when printed – document, written in a consumer-friendly manner, and focused on key summary information to be provided precontractually, before the retail investor makes a commitment to invest in a PRIIP. The KID is not designed to be exhaustive, but a summary document. The PRIIPs Regulation set out a number of features of the KID – notably, the sections to be contained, the headings of these sections and their order, and the information to be included in these sections – but left details of this information and its presentation to regulatory technical standards (RTS). The ESAs have been working since 2014 on finalising proposals for these RTS. The draft RTS, in keeping with the aims of the PRIIPs Regulation, have been prepared to offer a sound basis for comparing and understanding a wide range of PRIIPs. While the draft RTS cover a range of different areas of the KID, as determined under the PRIIPs Regulation, the core focus of the draft RTS is on the information related to risks, and a risk indicator in particular, the information related to rewards or performance, and the information related to costs. These are all areas where the variety and technical complexity of PRIIPs is such that simplifying information provided through summary indicators was foreseen as necessary already in the PRIIPs Regulation itself. These are also areas where developing such summary indicators for the wide range of PRIIPs involved, in view of their often significantly different features as products, was identified early on as a key challenge by the ESAs. Use of consumer testing In developing the draft RTSs, the ESAs have very specifically sought to reflect evidence from targeted consumer testing on the challenges faced by retail investors in using, understanding and comparing information on investment products, including in particular information on the risks, rewards (performance) and costs of these products. A study was conducted with the European Commission, working in collaboration with the ESAs, which showed that simpler presentations of information (such as different ways of showing risks, rewards and costs) were in general terms more effective for retail investors. Notably this included a simple risk indicator with a single risk scale with a limited
number (seven) of discrete ‘buckets’, a tabular presentation of reward or performance information, and tabular presentations of cost information. Summary Risk Indicator As mentioned, one of the key challenges faced in preparing the draft RTS was summarising a wide range of different risk profiles within a single ‘summary risk indicator’ (SRI). The ESAs have identified market and credit risk as the major factors of risk that need to be reflected in this indicator, alongside liquidity risk.

7
Measuring the market and credit risks in a way that allows for fair and objective comparisons between all the various PRIIPs on offer in the EU has required considerable expert work. On the market risk side, a method has been developed that is practical for products with mostly normal distributions of returns, but which is also able to address products with non-normal distributions, for instance with guarantees or capital protection of varying levels or other discontinuities in their pay out structures including precipice structures. The method has also been developed so as to reflect down-side risk, in view of the importance of this for retail investors. In order to allow for a reasonable comparison between different PRIIPs, the approach is benchmarked against the risk-free rate of return. For measuring credit risk, there is no generally applicable method that is available across the many different PRIIPs and their varying legal structures, but the use of credit ratings, supported in general by consultation responses, has been viewed as the most proportionate approach at this time, balancing a degree of objectivity with practicality. However, to properly take into account the impact of different legal structures, mitigation factors needed to be added, also in view of ensuring a fair treatment of those PRIIPs or PRIIP manufacturers that do not have a credit rating. The ESAs note that the measurement of credit risk in an objective way in the context of regulatory and prudential requirements has been a matter of some development and discussion over recent years, including a general policy of the European Commission to reduce reliance on credit ratings in EU law, and so believes the treatment of credit risk will need to be monitored, and may need to be further developed in the coming years, so as to build on these developments and discussions in the fairest and most objective way in view of the range of PRIIPs. Performance scenarios Performance information is linked closely with risk. Typically, higher performance comes at the price of higher risks (this may be true for both market and credit risk). Higher risks might lead to higher performance, but there are no guarantees of this. In this respect it is important to show information about performance. Ensuring this information is sufficiently harmonised and comparable is rather challenging, as future performance is of course not certain. However, after examining many options and assessing consultation responses, the ESAs have concluded in favour of using the measure of market risk as a basis for identifying a future spread of possible outcomes, this time without being benchmarked against the risk-free rate, but instead reflecting typical past returns over a suitable performance window. The aim in this approach is to show the spread or range of outcomes, and not to give undue certainty to these outcomes. In addition, consumer testing showed some challenges with using more graphical presentation
techniques, so a relatively simple use of tables, showing figures without including a probability of their achievement, was selected. Costs The work on cost disclosures focused on overall options for enhancing the comparability and transparency (comprehensibility for consumers) of the cost information to be included in the KID, and on progressing the technical underpinnings needed to obtain such information in an objective and comparable way between different PRIIPs. There are significant challenges given the scope of the products covered by the PRIIPs Regulation, in that for many of these products there has not been so far a standardised EU approach to the identification and disclosure of costs.

8
However, progress made for the KID can be expected to have a major positive impact, to galvanise further and significant improvements in transparency in an area that has been of major importance in recent years for regulators across the EU. Cost structures can be complex and their impact can vary significantly according to how long a consumer invests and how much they invest. What needs to be communicated is not only the overall level of costs, but how these costs can vary, and their relevant breakdown. However, providing too much information and too many figures can confuse consumers, who struggle to identify which of the many figures they see are relevant and important for them. The variety of the cost structures can also make it difficult for consumers to compare different costs between products. To address these comparability and comprehension challenges, the KID includes, pursuant to the PRIIPs Regulation, information on costs that:
 Summarises the overall impact of the costs, in monetary and percentage terms; and shows how the costs accumulate for different holding periods. The ESAs have considered that the use of a Reduction in Yield indicator would be the most appropriate for that purpose.  Identifies the key costs of the cost structure in a summary breakdown of costs, to enable a consumer to see how these might apply to how they might use the product. The ESAs have therefore detailed the different types of costs for the different types of PRIIPs, suggesting different methodologies to calculate or estimate these different types of costs, which can be explicit or implicit. Special cases The draft RTSs recognise certain special cases that arise. The first is with certain standardised exchange-traded derivatives, where a different approach on performance information was considered to offer more useful information for retail investors. Further to this, these PRIIPs highlighted the significant impact that updating and revising the KID might have. The requirements for the KID have therefore been carefully calibrated to ensure they do not need updating continuously. The KID must not be misleading or inaccurate, but the KID by its very nature as a pre-contractual document must contain information that is to a degree generic or general in nature. The KID is designed, in this regard, to show example investments. The second is PRIIPs offering multiple investment options. The KID for these can either be a generic one related to the product in general, which would include information on the range of risks and costs, but not all the detail on every specific option, or specific KIDs can be produced for each option to combine the generic information with the specific information for that option. In the case of the former (generic KID), further information needs to be provided, under the responsibility of the PRIIP manufacturer, on the specific options offered, and this information needs to be consistent with the information that would be in the KID on risks, performance and costs. There is, however, no requirement to reflect every possible combination of underlying options, which may be very large in view of the structure of some PRIIPs – for instance, the information provided would not need to reflect specific choices for a certain proportion and selection of different underlying investment options. This ‘personalisation’ of the information to reflect the

9 details of a specific transaction is not required; the KID is aimed at providing pre-contractual information to aid the retail investor in considering and comparing options before they have made an investment decisions, not at illustrating the investment decision they have already made. Impacts The ESAs are conscious that implementing the PRIIPs Regulation and the draft RTS will have significant impacts for different industry sectors. However, the ESAs believe that these impacts – both those following from the PRIIPs Regulation itself, which is the main driver of impacts, and those that are driven by the specific measures foreseen in these daft RTS – are justified in view of the consumer protection outcomes being sought and the necessity of measures to improve transparency, comprehension and comparability in the retail investment markets in the EU. The major ‘cost drivers’ the ESAs have identified in respect of the specific methodologies selected are those related to putting in place the relevant systems for gathering data for the calculation of the market risk measure (MRM), the credit risk measure (CRM), the performance scenario information, and the summary figures for the costs, including on portfolio transaction costs for funds. In the absence of the draft RTS, the obligations in the PRIIPs Regulation – which clearly require summary figures – would, however, in the view of the ESAs, have raised significant cost implications in any case, whilst a lack of consistency across sectors and national markets would significantly reduce the effectiveness of the PRIIPs KID for consumers. To reduce costs of implementation and aid the development of the KID by PRIIP manufacturers, the ESAs plan to develop detailed guidance on the methods to follow for the calculation of the SRI, the performance scenarios, and the cost information to be included in the KID. The ESAs also plan to issue detailed ‘FAQs’ to aid in interpretation, and to develop supporting level three guidelines and guidance where this is needed in view of supervisory convergence and easing implementation.

EN
3. Draft RTS
EUROPEAN
COMMISSION
Brussels, XXX
… XXX draft
COMMISSION DELEGATED REGULATION (EU) No …/.. of XXX on […]

11
COMMISSION DELEGATED REGULATION (EU) …/.. of XXX supplementing Regulation (EU) No 1286/2014 of the European Parliament and of the Council on key information documents for packaged retail and insurance-based investment products (PRIIPs) by laying down regulatory technical standards with regard to the presentation, content, review and revision of key information documents and the conditions for fulfilling the requirement to provide such documents THE EUROPEAN COMMISSION, Having regard to the Treaty on the Functioning of the European Union, Having regard to Regulation (EU) No 1286/2014 of the European Parliament and of the Council of 26 November 2014 on key information documents for packaged retail and insurance-based investment products (PRIIPs) 3 , and in particular Article 8(5), Article 10(2) and Article 13(5) thereof, Whereas:
(1) Regulation (EU) No 1286/2014 introduces a new standardised key information document to improve the retail investor's understanding of packaged retail and insurance-based investment products (‘PRIIPs’) and the comparability of those products. (2) In order to provide retail investors with key information that is easy to read, understand and compare, a common template should be established for the key information document. (3) The identity and contact details mentioned in Article 8(3)(a) of Regulation (EU) No 1286/2014 should include the International Securities Identification Number or Unique Product Identifier for the PRIIP where this is available in order to make it easier for the retail investor to find additional information about the PRIIP. (4) In order to ensure that retail investors understand and compare the economic and legal features of the PRIIP, as well as to provide them with an appropriate overview of the investment policy and strategy of the PRIIP, the key information document should contain standardised information concerning the type of the PRIIP, its investment objectives and how they will be achieved and the key features or aspects of the product, such as the insurance coverage. (5) The information provided to retail investors should enable those investors to understand and compare the risks associated with investments in PRIIPs so that they can make informed investment decisions. The risks pertaining to a PRIIP can

3
OJ L352, 9.12.2014, p.1.

12 vary; the most important risks can be classified as market risk, credit risk and liquidity risk. In order for retail investors to fully understand those risks, information on the risks should be aggregated as far as possible and numerically presented as a single summary risk indicator with sufficient narrative explanations. (6) When assessing credit risk, PRIIP manufacturers should be able to take into account certain factors that may mitigate the credit risk for a retail investor. In this respect, where assessing that the assets of a PRIIP or appropriate collateral, or assets backing the payment obligations of a PRIIP, are at all times until maturity equivalent to the payment obligations of the PRIIP to its investors, this should reflect the assets held by an insurance undertaking so as to correspond at any time to the current amount that the insurance undertaking would have to pay to transfer its obligations in respect of the PRIIP to another insurance undertaking. (7) The ESAs recognise that currently ECAI ratings provide a consistent proxy for credit risk across different EU sectors, although it is general Commission policy to reduce the reliance on credit ratings wherever possible. In view of this, and the importance of the objectivity, accuracy and comparability of the summary risk indicator foreseen in this Regulation, the summary risk indicator and the contribution to this of market and credit risk should be monitored, so that evidence on their effectiveness in practice can be made available for the review of Regulation (EU) No 1286/2014 foreseen by 31 December 2018, This should take into account the extent to which ECAI ratings in practice reflect the creditworthiness of the PRIIP manufacturer and the credit risk faced by investors in individual PRIIPs. (8) While estimates on returns from a PRIIP are difficult to produce and understand, information on such estimates are of primary interest for the retail investors and should be included in the key information document. Retail investors should be provided with clear information on return estimates that is consistent with realistic assumptions about possible outcomes, with the estimates of the PRIIPs’ level of market risk, presented in such a way as to make clear the uncertainty of that information and the fact that better or worse outcomes are possible. (9) In order for the retail investor to be able to appreciate the risk, the key information document should provide the retail investor with information as to potential consequences where a PRIIP manufacturer is not able to pay out. The degree of protection of the retail investor in such cases under investment, insurance or deposit guarantee schemes should be clearly set out. (10) Information on costs is important for retail investors when comparing different PRIIPs, which can have different cost structures, and when considering how the cost structure of a particular PRIIP might apply to them, which depends on how long they are
invested, how much they invest, and how well the PRIIP performs. For this reason, the key information document should contain information that allows the retail investor to compare the overall total cost levels between different PRIIPs when held for their recommended holding periods and shorter periods, and to understand how these costs might vary and evolve over time.

13
(11) Consumer testing research has shown that retail investors can understand monetary figures more readily than percentages. Small differences in costs expressed in percentages may correlate with large differences in the costs borne by the retail investor when expressed in monetary terms. For this reason, the key information document should also provide the total costs for the recommended holding periods and shorter periods in monetary terms as well as in percentages. (12) Given that the impact of different kinds of costs can vary according to how long a retail investor keeps his investment in a PRIIP, the key information document should also provide a breakdown of the different costs that make up the total costs. The breakdown of costs should be expressed in standardised terms so that the amounts for different PRIIPs can be easily compared. So that they can be easily applied to different investment amounts than those used to calculate the total costs in monetary terms, the breakdown of costs should be expressed in percentage terms. (13) Retail investors may experience a change in personal circumstances where longer term investments unexpectedly need to be disinvested. Disinvestments due to market developments may also be necessary. Given the difficulties for retail investors to anticipate the degree of liquidity they may need in their investment portfolios as a whole, information on recommended holding periods and required minimum holding periods, and the possibility of partial or complete early exit, is particularly important for them and should be included in the key information document. For the same reasons, the availability and consequences of such early disinvestment should be made clear. Specifically, it should be clear whether such consequences are due to explicit fees, penalties or limitations on disinvestment rights, or to the fact that the value of the particular PRIIP to be disinvested is particularly sensitive to the timing of the disinvestment. (14) Given that the key information document is also likely to be used as a summary of the main features of the PRIIP by retail investors, it should contain clear information on how a complaint might be lodged about the product or about the conduct of the PRIIP manufacturer or a person advising on, or selling, the product. (15) Some retail investors may wish to obtain further information on specific aspects of the PRIIP. The key information document should therefore include a clear and specific cross-reference to where further specific information can be found, where such information is to be included in the key information document pursuant to Regulation (EU) No 1286/2014. Where the PRIIP manufacturer is obliged to disclose certain other information according to national or Union law the retail investor should be informed of this fact and of how to obtain those other documents, even if they are only to be provided on request. In view of ensuring that the key information document is as
concise as possible, links to those other documents may be provided by means of a website, as long as their existence is made clear and they can be accessed by means of that website. (16) A key information document for a PRIIP that offers many underlying investment options cannot be provided in the same format as a key information document for

14 another PRIIP, since each underlying investment option will have a specific risk, performance and cost profile, which prevents all necessary information to be provided in a single, concise stand-alone document. The underlying investment options may be investments in PRIIPs or other investments of a similar nature, or standardised portfolios of underlying investments. Each of these underlying investment options would have, by its nature, different risks, rewards and costs. Depending on the nature and number of underlying investment options the PRIIP manufacturer should therefore, if appropriate, be able to prepare individual key information documents for each option, including information within these key information documents about the PRIIP in general. (17) Where this is not appropriate for the retail investors, the PRIIP manufacturer should provide the specific information about the underlying investment options and the generic information about the PRIIP separately. To avoid confusion, the generic information about the PRIIP provided in the key information document should indicate the range of risks, performance and costs that can be expected across the different underlying investment options offered. In addition, the specific information on the underlying investment options should always reflect the features of the PRIIP through which the PRIIP manufacturer is offering the underlying investment options. This specific information may be provided in different forms, for example in the form of single document setting out the necessary information on all the different underlying investment options, or through individual documents for each underlying investment option. Regardless of the form chosen, the specific information should always be consistent with the information that is contained in a key information document, for instance include consistent summary risk indicators and cost disclosures. (18) PRIIP manufacturers must prepare key information documents that are accurate, fair, clear and not misleading. The information contained in the document should be capable of being relied on by a retail investor when making an investment decision, even in the months and years following the initial preparation of the key information document, for those PRIIPs that remain available to retail investors, Standards should therefore be laid down to ensure timely and appropriate review and revision of key information documents, so that the key information documents provided to retail investors remain accurate, fair, clear and not misleading. (19) Data that is used for preparing the information contained in the key information document, such as data on costs, risks and performance scenarios, may change over time. Changing data can lead to changes in the information to be included, such as a change in the risk or costs indicators. For this reason, PRIIP manufacturers should put in place periodic processes to review the information contained in the key
information document. Those processes should include an assessment of whether changes in the data would necessitate a revision and republication of the document. The approach by PRIIP manufactures should reflect the extent to which the information to be included in the key information document changes, for instance for an exchange-traded derivative, such as a standardised future, call or put, there should be no necessity to continuously update the key information document as the

15 information required for these instruments on their risks, rewards and costs would not fluctuate. (20) Periodic reviews may not be sufficient in cases where the PRIIP manufacturer becomes aware or should have become aware of changes outside the periodic review process that may significantly impact the information contained in the key information document, such as changes to a previously disclosed PRIIP investment policy or strategy that would be significant for retail investors, or significant changes to the cost structure or risk profile. For this reason, PRIIP manufacturers should also be required to establish processes for identifying situations where the information contained in the key information document should be reviewed and revised on an ad hoc basis. (21) Where a periodic or ad hoc review of a key information document identifies changes to the information that is required to be included in the document, or concludes that information contained in the key information document is no longer accurate, fair, clear and not misleading, the PRIIP manufacturer should be required to revise the key information document to take that changed information into account. (22) Given that changes may be relevant for retail investors and their future allocation of investment assets, the retail investors should be able to easily locate the new key information document, which should therefore be published, and be clearly identifiable, on the website of the PRIIP manufacturer. Where it is possible and the PRIIP manufacturer has access to the details of retail investors, the PRIIP manufacturer should inform the retail investors when the key investor documents have been revised, for example by means of mailing lists or email alerts. (23) In order to ensure that the timing of the delivery of key information documents is approached in a consistent way across the Union, the PRIIP manufacturers should be required to provide the key information document in good time before those retail investors are bound by any contract or offer relating to that PRIIP. (24) The key information document should be made available to the retail investors sufficiently prior to their investment decision, so that they are able to understand and take into account the relevant PRIIP information when making that decision. Since the investment decision is made prior to the commencement of any mandatory cooling off period, the key information document should be provided prior to such a cooling off period. (25) While in all cases retail investors should receive the key information document in good time before they are bound by any contract or offer related to the PRIIP, what might be considered sufficient time for the retail investor to understand and take into account the information may vary, given that different retail investors have different needs, experience and knowledge. The person advising on, or selling, a PRIIP should therefore take into account such factors in relation to
individual retail investors when establishing the time that those retail investors will need to consider the contents of the key information document.

16
(26) In order to make an informed investment decision, a retail investor may need additional time to consider the key information document of a complex PRIIP or a PRIIP that is unknown to that investor. Accordingly, such factors should be taken into account when considering what amounts to the provision of the key information document in good time. (27) The urgency of the situation, for instance where it is important for a retail investor to buy a PRIIP at a given price and the price is sensitive to the timing of the transaction, should also be considered when determining the extent of the good time criterion. (28) This Regulation is based on the draft regulatory technical standards submitted to the Commission by the European Banking Authority, the European Insurance and Occupational Pensions Authority and the European Securities and Markets Authority (the ‘European Supervisory Authorities’). (29) The European Supervisory Authorities have conducted open public consultations on the draft regulatory technical standards on which this Regulation is based, analysed the potential related costs and benefits and requested the opinion of the Banking Stakeholder Group established in accordance with Article 37 of Regulation (EU) No 1093/2010 of the European Parliament and of the Council 4 , the Insurance and Reinsurance Stakeholder Group established in accordance with
Article 37 of Regulation (EU) No 1094/2010 of the European Parliament and of the
Council 5
, and the Securities and Markets Stakeholder Group established in accordance with Article 37 of Regulation (EU) No 1095/2010 of the European Parliament and of the Council 6 , HAS ADOPTED THIS REGULATION:
CHAPTER I
Content and presentation of the key information document
Article 1
General information section
The section in the key information document that relates to the identity of the PRIIP manufacturer and its competent authority shall contain all of the following information:

4
Regulation (EU) No 1093/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Banking Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/78/EC (OJ L 331, 15.12.2010, p. 12). 5 Regulation (EU) No 1094/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Insurance and Occupational Pensions Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/79/EC (OJ L 331, 15.12.2010, p. 48). 6 Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/77/EC (OJ L 331, 15.12.2010, p. 84).

17
(a) the name of the PRIIP assigned by the PRIIP manufacturer and, where present, the PRIIP´s International Securities Identification Number or Unique Product Identifier; (b) the legal name of the PRIIP manufacturer; (c) the PRIIP manufacturer’s specific website address providing retail investors with information on how to get in contact with the PRIIP manufacturer, and a telephone number; (d) the name of the competent authority responsible for the supervision of the PRIIP manufacturer in relation to the key information document; (e) either the date of production or the date of the latest revision of the key information document.
Article 2
‘What is this product?’ section

Information relating to the type of the PRIIP in the section entitled ‘What is this
product?’ of the key information document shall describe its legal form.
Information stating the objectives of the PRIIP and the means for achieving those
objectives in the section entitled ‘What is this product?’ of the key information document shall be summarised in a brief, clear and easily understandable manner. That information shall identify the main factors upon which return depends, the underlying investment assets or reference values, and how the return is determined, and the relationship between the PRIIP’s return and that of the underlying investment assets or reference values. Where relevant, including in the case of long term PRIIPs, the information shall reflect the relationship between the recommended holding period and the risk and reward profile of the PRIIP. Where the number of assets or reference values referred to in the first subparagraph is such that specific references to all of them are not possible in view of the length of the key information document, the information may only identify market segments or instrument types in respect of underlying investment assets or reference values.
The description of the type of retail investor to whom the PRIIP is intended to be
marketed in the section entitled ‘What is this product?’ of the key information document shall include information on the target retail investors identified by the PRIIP manufacturer in particular depending on the ability of the retail investors to bear investment loss and their investment horizon preferences.
The details of insurance benefits in the section entitled ‘What is this product?’ of the
key information document shall include, in a general summary, the key features of the insurance contract, the definition of each benefit included, and example information, which is consistent with the remainder of the key information document, showing the overall premium, the biometric risk premium that forms part of this overall premium and the amount invested, in the case where the premium is paid in the form of a single

18 lump sum, and, in the case the where the premium is paid in a period form, the number of periodic payments, an estimation of the average biometric risk premium as a percentage of the annual premium, and an estimation of the average amount invested.
5. The information relating to the term of the PRIIP in the section entitled ‘What is this
product?’ of the key information document shall include all of the following:
(a) the maturity date of the PRIIP or an indication that there is no maturity date; (b) whether the PRIIP manufacturer is entitled to terminate the PRIIP unilaterally; (c) a description of the circumstances under which the PRIIP can be automatically terminated, and the termination dates, if known.
Article 3
‘What are the risks and what could I get in return?’ section

In the section entitled ‘What are the risks and what could I get in return?’ of the key
information document, PRIIP manufacturers shall apply the methodology for the presentation of risk as set out in Annex II, shall include the technical aspects for the presentation of the summary risk indicator as set out in Annex III and shall comply with the technical guidance, formats and the methodology for the presentation of performance scenarios, as set out in Annexes IV and V.
In the section entitled ‘What are the risks and what could I get in return?’ of the key
information document, PRIIP manufacturers shall include the following:
(a) the level of risk of the PRIIP in the form of a risk class by using a summary risk indicator having a numerical scale from 1 to 7; (b) an explicit reference to any illiquid PRIIP or PRIIP with materially relevant liquidity risk, as defined in Part 4 of Annex II, in a warning to this effect in the presentation of the summary risk indicator; (c) a narrative below the summary risk indicator explaining to the retail investor that if a PRIIP is denominated in a currency other than the applicable currency of the legal tender of the Member State where the PRIIP is being marketed, the return the retail investor gets, when expressed in the currency of the Member State where the PRIIP is being marketed, may change depending on currency fluctuations; (d) a brief description of the PRIIP´s risk and reward profile, where this is the case, a warning that the risk of the product if not held to maturity or for the recommended holding period may be significantly higher than the one represented in the summary risk indicator; (e) for PRIIPs with contractually agreed-upon early exit penalties or long disinvestment notice periods, a reference to the relevant underlying conditions in the section ‘How long should I hold it and can I take money out early?’.

19
(f) an indication of the possible maximum loss, and that the investment may be lost if it is not protected or if the PRIIP manufacturer is unable to pay out, or that necessary additional investment payments may be required to the initial investment and the total loss may significantly exceed the total initial investment.
3. PRIIP manufacturers shall include three appropriate performance scenarios, as set out
in Annex V in the section entitled ‘What are the risks and what could I get in return?’ of the key information document. Those three performance scenarios shall represent an unfavourable scenario, a moderate scenario and a favourable scenario.
4. For PRIIPs where the PRIIP manufacturer considers that significant risks of loss are
not adequately covered by the three appropriate performance scenarios, the PRIIP manufacturer may include an additional performance scenario in the section entitled ‘What are the risks and what could I get in return?’ of the key information document.
5. For insurance-based investment products, an additional performance scenario shall be
included in the section entitled ‘What are the risks and what could I get in return?’ of the key information document reflecting the insurance benefit the beneficiary receives where a covered insurance event occurs.
6. For PRIIPs that are exchange-traded derivatives that are futures, call options and put
options traded on a regulated market or on a third-country market considered to be equivalent to a regulated market in accordance with Article 28 of Regulation (EU) 600/2014, and as such do not fall within the definition of an OTC derivative as defined in Article 2(7) of Regulation (EU) 648/2012, performance scenarios shall be included in the form of pay-off structure graphs as set out in Annex V in the section entitled ‘What are the risks and what could I get in return?’ of the key information document.
Article 4
‘What happens if [the name of the PRIIP manufacturer] is unable to pay out?’ section PRIIPs manufacturer shall include the following in the section entitled ‘What happens if [the name of the PRIIP manufacturer] is unable to pay out?’ of the key information document:
(a) an indication whether the retail investor may face a financial loss due to the default of the manufacturer or to the default of an entity other than the PRIIP manufacturer, and the identity of that entity; (b) a clarification whether the loss referred to in point (a) is covered by an investor compensation or guarantee scheme, and whether there are any limitations or conditions to that cover.
Article 5
‘What are the costs?’ section

PRIIP manufacturers shall apply the following in the section entitled ‘What are the
costs?’ of the key information document:

20
(a) the methodology for the calculation of costs set out in Annex VI; (b) the ‘Costs over time’ and ‘Composition of costs’ tables to information on costs, as set out in Annex VII in accordance with the relevant technical guidance therein.
2. In the ‘Costs over time’ table in the section entitled ‘What are the costs?’ of the key
information document, PRIIP manufacturers shall specify the summary cost indicator of the total aggregated costs of the PRIIP as a single number in monetary and percentage terms for the different time periods set out in Annex VI.
3. In the ‘Composition of costs’ table in the section entitled ‘What are the costs?’ of the
key information document, PRIIP manufacturers shall specify the following:
(c) any one-off costs, as entry and exit costs, and present them in percentage terms; (d) any recurring costs, as portfolio transaction costs per year, insurance costs – where relevant, and other recurring costs per year, and present them in percentage terms; (e) any relevant incidental costs, such as performance fees or carried interest, and present them in percentage terms.
4. PRIIP manufacturers shall insert a description of each of the different costs included in
the ‘Composition of costs’ table in the section entitled ‘What are the costs?‘ of the key information document, specifying where and how such costs may differ from the actual costs the retail investor may incur, where the costs have been estimated in accordance with Annex VI, or where the specific level of certain costs may depend on the retail investor choosing to exercise or not exercise certain options. For insurancebased investment products, the PRIIP manufacturer shall insert below the ‘Composition of costs’ table a narrative explaining the impact of the insurance benefits of the product on the returns of the investment for the retail investor.
Article 6
‘How long should I hold it and can I take my money out early?’ section PRIIP manufacturers shall include the following in the section entitled ‘How long should I hold it and can I take my money out early?’ of the key information document:
(a) a brief description of the reasons for the selection of the recommended holding period and, where present, the required minimum holding period; (b) a description of the features of the disinvestment procedure and when disinvestment is possible, including an indication of the impact of cashing-in early on the risk or performance profile of the PRIIP, including whether cashing-in early has an impact on the applicability of capital guarantees; (c) information about any fees and penalties which are incurred for disinvestments prior to maturity, including a cross reference to the information on costs to be included in the key information document pursuant to Article 5, including a clarification of the impact of such fees and penalties for different holding periods.

21
Article 7
‘How can I complain?’ section
PRIIP manufacturers shall provide the following information in the section entitled ‘How can I complain?’ of the key information document, in summary format:
(a) steps to be followed for lodging a complaint about the product or the conduct of the PRIIP manufacturer or the person advising on, or selling, the product; (b) a link to the relevant website for such complaints; (c) an up-to-date postal address and an email address to which such complaints may be submitted.
Article 8
‘Other relevant information’ section

PRIIP manufacturers shall indicate any additional information documents that may be
provided in the section entitled ‘Other relevant information’ of the key information document, and whether such additional information documents are mandatorily made available under Union or national law, or only available at the request of the retail investor.
The information included in the section entitled ‘Other relevant information’ of the
key information document may be provided in summary format, including a link to the website where further details to documents are made available.
Article 9
Template
PRIIP manufacturers shall present the key information document by means of the template laid down in Annex I, and complete it in accordance with the requirements set out in this Regulation and in Regulation (EU) No 1286/2014.
CHAPTER II
Specific provisions on the key information document
Article 10
PRIIPs offering a range of options for investment Where a PRIIP offers a range of underlying investment options, and the information regarding those underlying investment options cannot be provided within a single, concise, stand-alone key information document, PRIIP manufacturers shall produce one of the following:
(a) a key information document for each underlying investment option within the PRIIP, including information about the PRIIP overall, in accordance with
Chapter I;

22
(b) a generic key information document describing the overall PRIIP in accordance with Chapter I unless otherwise specified in Articles 11 to 14.
Article 11
‘What is this product’ section in the Generic key information document In the section entitled ‘What is this product’, by way of derogation from paragraphs 2 and 3 of Article 2, PRIIP manufacturers shall specify the following:
(a) a description of the types of underlying investment options, including the market segments or instrument types, as well as the main factors upon which return depends; (b) a statement indicating that the type of investors to whom the PRIIP is intended to be marketed varies on the basis of the underlying investment option; (c) an indication where the specific information on each underlying investment option is to be found.
Article 12
‘What are the risks and what could I get in return?’ section in the Generic key information document In the section entitled ‘What are the risks and what could I get in return?’, by way of derogation from paragraphs 2(a) and 3 of Article 3, PRIIP manufacturers shall specify the following:
(a) the range of risk classes of all underlying investment options offered within the PRIIP by using a summary risk indicator having a numerical scale from 1 to 7, as set out in Annex III; (b) a statement indicating that the risk and return of the investment varies on the basis of the underlying investment option; (c) a brief description on how the performance of the PRIIP as a whole depends on the underlying investment options; (d) an indication where the specific information on each underlying investment option is to be found.
Article 13
‘What are the costs?’ section in the Generic key information document In the section entitled ‘What are the costs?’, by way of derogation from article 5(1)(b), PRIIP manufacturers shall specify the following:
(a) the range of costs for the PRIIP in the 'Costs over time' and 'Composition of costs' tables set out in Annex VII;

23
(b) a statement indicating that the costs to the retail investor vary on the basis of the underlying investment option; (c) an indication where the specific information on each underlying investment option is to be found.
Article 14
Specific information on each underlying investment option In relation to the specific information referred to in Articles 11, 12 and 13, PRIIP manufacturers shall include for each underlying investment option – all of the following:
(a) a comprehension alert, where relevant;
(b) the investment objectives, the means for achieving them, and the intended target market in accordance with paragraphs 2 and 3 of Article 2; (c) a summary risk indicator and narrative, and performance scenarios in accordance with Article 3; (d) a presentation of the costs in accordance with Article 5.
CHAPTER III
Review and revision of the key information document
Article 15
Review

PRIIP manufacturers shall review the information contained in the key information
document every time there is a change that significantly affects or is likely to affect significantly the information contained in the key information document and, at least, every twelve months following the date of the initial publication of the key information document.
The review referred to in paragraph 1 shall verify whether the information contained
in the key information document remains accurate, fair, clear, and non-misleading. In particular, it shall verify:
(a) whether the information contained in the key information document is/remains compliant with the general form and content requirements under Regulation (EU) No 1286/2014, or with the specific form and content requirements laid down in this Regulation; (b) whether the PRIIP´s market risk or credit risk measures have changed, where such a change has the combined effect that necessitates the PRIIP´s move to a different class of the summary risk indicator from that attributed in the key information document subject to review; (c) whether the mean return for the PRIIP’s moderate performance scenario, expressed as an annualised percentage return, has changed by more than five percentage points.

24
3. For the purposes of paragraph 1, PRIIP manufacturers shall establish and maintain
adequate processes throughout the life of the PRIIP where it remains available to retail investors to identify without undue delay any circumstances which might result in a change that affects or is likely to affect the accuracy, fairness or clarity of the information contained in the key information document.
Article 16
Revision

PRIIP manufacturers shall without undue delay revise the key information document
where a review pursuant to Article 15 concludes that changes to the key information document need to be made.
PRIIP manufacturers shall ensure that all sections of the key information document
affected by such changes are updated.
The PRIIP manufacturer shall publish the revised key information document on its
website.
CHAPTER IV
Delivery of the key information document
Article 17
Conditions on good time
The person advising on or selling, a PRIIP shall provide the key information document sufficiently early so as to allow retail investors enough time to consider the document before being bound by any contract or offer relating to that PRIIP, and disregarding whether or not the retail investor is provided with a cooling off period. To this end, the person advising on or selling a PRIIP shall assess the time needed by each retail investor to consider the key information document, taking into account the following:
(a) the knowledge and experience of the retail investor with the PRIIP or PRIIPs of a similar nature or with risks similar to those arising from the PRIIP; (b) the complexity of the PRIIP; (c) where the advice or sale is at the initiative of the retail investor, the urgency explicitly expressed by the retail investor of concluding the proposed contract or offer.
Article 18
Final Provision
This Regulation shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union. This Regulation shall be binding in its entirety and directly applicable in all Member States. Done at Brussels …

25
For the Commission

26
ANNEX I
TEMPLATE OF THE KEY INFORMATION DOCUMENT
PRIIP manufacturers shall comply with the section order and titles set out in the template, which however does not fix parameters regarding the length of individual sections and the placing of page breaks, and is subject to an overall maximum of three sides of A4-sized paper when printed.

29
ANNEX II
METHODOLOGY FOR THE PRESENTATION OF RISK
PART 1
Market risk assessment
Determination of the market risk measure (MRM)

Market risk is measured by the annualised volatility corresponding to the value-at-risk
(VaR) measured at the 97.5% confidence level over the recommended holding period unless stated otherwise. The VaR is the percentage of the amount paid that is returned to the retail investor.
The PRIIP shall be assigned a market risk measure (MRM) class according to the
following table:
MRM class Annualised volatility (VEV)
1 < 0.5 %
2 0,5 % - 5.0 %
3 5.0 %-12 %
4 12 %-20 %
5 20 %-30 %
6 30 %-80 %
7 >80 %
Specification of PRIIP categories for the purposes of the market risk assessment
For the purposes of determining market risk, PRIIPs are divided into four categories.
Category 1 covers the following:
(a) PRIIPs where investors could lose more than the amount they invested; (b) PRIIPs that fall within one of the categories referred to in [items 4 to 10 of Section C of Annex I to Directive 2014/65/EU of the European Parliament and of the Council 7 and similar instruments issued as transferable securities]; (c) PRIIPs or the underlyings of these having prices on a less regular basis than monthly, or which do not have an appropriate benchmark or proxy, or whose appropriate benchmark or proxy has prices on a less regular basis than monthly.
Category 2 covers PRIIPs, which either directly or on a synthetic basis, offer nonleveraged exposure to the prices of underlying investments, or a leveraged exposure

7
Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU (OJ L 173, 12.6.2014, p. 349).

30 on underlying investments that pays a constant multiple of the prices of these underlying investments, where at least 2 years of historical daily prices, or 4 years of historical weekly prices, or 5 years of monthly prices are available for the PRIIP, or where existing appropriate benchmarks or proxies are available, provided that such benchmarks or proxies have prices which are as regular as such prices referred to above.
6. Category 3 covers PRIIPs whose values reflect the prices of underlying investments,
but not as a constant multiple of the prices of these underlying investments, where at least 2 years of daily prices of the underlying assets, 4 years of weekly prices or 5 years of monthly prices, or where existing appropriate benchmarks or proxies are available, provided that such benchmarks or proxies have prices which are as regular as such prices referred to above.
7. Category 4 covers any PRIIP whose value depends in part on factors not observed in
the market. This includes insurance-based PRIIPs which distribute to retail investors a portion of the PRIIP manufacturer’s profits. Use of appropriate benchmarks or proxies Where appropriate benchmarks or proxies are used by a PRIIP manufacturer, these shall be representative of the assets or exposures that determine the performance of the PRIIP. The PRIIP manufacturer will document the use of such benchmarks or proxies. MRM class determination for category 1 PRIIPs
8. The MRM class for category 1 PRIIPs shall be 7, with the exception of those referred
to in point 4 (c), where the MRM class shall be 6.
MRM class determination for category 2 PRIIPs
9. The VaR shall be calculated from the moments of the observed distribution of returns
of the PRIIP’s or its benchmark or proxy’s price during the past 5 years. The minimum frequency of observations is monthly. When prices are available on a daily basis, the frequency shall be daily; when prices are available on a weekly basis, the frequency shall be weekly; when prices are available on a bi-monthly basis, the frequency shall be bi-monthly.
10. When 5 years of daily prices is not available, a shorter period may be used. For daily
observations of a PRIIP’s or its benchmark or proxy’s price, there shall be at least 2 years of observed returns. For weekly observations of a PRIIP’s price, there shall be at least 4 years of observed data. For monthly observations of a PRIIP’s price, there shall be at least 5 years of observed data.

31
11. The return over each period is defined as the natural logarithm of the ratio of the price
at the market close at the end of the current period to the market close at the end of the preceding period.
12. The VaR measure in return space is given by the Cornish-Fisher expansion, as follows:
VaRRETURN SPACE =  √N * (-1.96 + 0.474 * µ1 / √N – 0.0687 * µ2 / N + 0.146 * µ1 2 / N) – 0.5 2N where N is the number of trading periods in the recommended holding period; and , 1, 2 are respectively the volatility, skew and excess kurtosis measured from the return distribution. The volatility, skew and excess kurtosis are calculated from the measured moments of the distribution of returns in accordance with the following:
 the zero moment, M0, is the count of the number of observations in the period as under point 10  the first moment, M1, is the mean of all the observed returns in the sample  the second (M2), third (M3) and fourth (M4) moments are defined in the standard manner:
M2 = ∑i(ri – M1)
2 /M0,
M3 = ∑i(ri – M1)
3 /M0,
M4 = ∑i(ri – M1)
4 /M0, where ri is the return measured on the i th period in the history of returns.  the volatility, , is given by √M2.  the skew, µ1, is equal to M3 /  3 .  the excess kurtosis, µ2, is equal to M4 /  4 – 3.
13. The VaR-equivalent volatility (VEV) is given by:
VEV = {√(3.842 – 2* VaRRETURN SPACE) -1.96} / √T where T is the length of the recommended holding period in years.
14. For PRIIPs that are managed according to investment policies and/or strategies that
pursue certain reward objectives by participating through flexible investment in different financial asset classes (e.g. in both equity and fixed-income markets), VEV that shall be used shall be:
(a) Where there has been no revision of the investment policy over the period under point 10, take the maximum of:
(i) the actual VEV computed according to points 9-13;

32
(ii) VEV of the returns of the pro-forma asset mix that is consistent with the reference asset allocation of the fund at the time of the computation; (iii) the VEV which is consistent with the risk limit of the fund, if any and appropriate. (b) Where investment policy has been revised over the period under point 10, take the maximum of (ii) and (iii) above.
15. The PRIIP shall be assigned to a MRM class as laid down under point 2 depending on
the VEV. In the case of a PRIIP having only monthly price data, the MRM class obtained under point 2 shall be increased by one additional class. MRM class determination for Category 3 PRIIPs
16. The VaR in price space shall be calculated from a distribution of PRIIP values at the
end of the recommended holding period. The distribution shall be obtained by simulating the price or prices, which determine the value of the PRIIP, to the end of the recommended holding period. The VaR shall be the value of the PRIIP at the 97.5% confidence level at the recommended holding period discounted to the present date using the expected risk-free discount factor from the present date to the end of the recommended holding period.
17. The VaR-equivalent volatility (VEV) is given by:
VEV = {√(3.842 – 2* ln(VaRPRICE SPACE)) -1.96} / √T where T is the length of the recommended holding period in years. Only in cases where the product is called or cancelled before the recommended holding period according to the simulation, the period in years until the call or cancellation is used in the calculation.
18. The PRIIP shall be assigned to a MRM class as laid down in point 2, depending on
this VEV. In the case of a PRIIP having only monthly price data, the MRM class obtained under point 2 shall be increased by one additional class.
19. The minimum number of simulations is 10,000.
20. The simulation is based on bootstrapping the expected distribution of prices or levels
for the PRIIP’s underlying contracts from the observed distribution of returns for these contracts with replacement.
21. For the purposes of the simulation, there are two types of market observables that may
contribute to a PRIIP’s value: spot prices (or levels) and curves.
22. For each simulation of a spot price or level, the PRIIP manufacturer shall:
(a) calculate the return for each observed period in the past 5 years, or the years as set out under point 6, by taking the logarithm of the price at the end of each period divided by the price at the end of the previous period;

33
(b) randomly select one observed period which identifies the return for all underlying contracts for each simulated period in the recommended holding period (the same observed period may be used more than once in the same simulation); (c) calculate the return for each contract by summing the returns from the selected periods and correcting this return to ensure that the expected return measured from the simulated distribution of returns is the risk-neutral expectation of the return over the recommended holding period. The final value of the return is given by:
Return = E[Returnrisk-neutral ]– E[Returnmeasured] - 0.5  2 N –  ccy N Where:
 The second term corrects for the impact of the mean of the observed returns;  The third term corrects for the impact of the variance of the observed returns;  The last term corrects for the quanto impact if the strike currency is different from the asset currency. The terms contributing to the correction are:
 is the correlation between the asset price and the relevant Fx rate measured over the recommended holding period;  is the measured volatility of the asset; ccy is the measured volatility of the Fx rate. (d) Calculate the price of each underlying contract by taking the exponential of the return.
23. For curves a principal component analysis (PCA) shall be performed to ensure that the
simulation of the movements of each point on the curve over a long period results in a consistent curve. (a)The PCA is performed by:
(i) collecting the historical record of tenor points that define the curve for each trading period over the past 5 years, or the years as set out under point 6; (ii) ensuring that each tenor point is positive; where there is a negative tenor point, all tenor points shall be shifted by the minimum whole number or percentage to ensure positive values for all tenor points; (iii) calculating the return over each period for each tenor point by taking the natural logarithm of the ratio between the price/level at the end of each observed period and the price/level at the end of the preceding period;

34
(iv) correcting the returns observed at each tenor point so that the resulting set of returns at each tenor point has a zero mean; (v) calculating the covariance matrix between the different tenors by summing over returns; (vi) calculate the eigenvectors and eigenvalues of the covariance matrix. (vii) selecting the eigenvectors that correspond to the three largest eigenvalues. (viii) forming a matrix with 3 columns where the first column is the eigenvector with the largest eigenvalue; the middle column is the eigenvector with the second-largest eigenvalue and the last column is the eigenvector with the third-largest eigenvalue. (ix) projecting the returns onto the 3 principle eigenvectors calculated in the previous step by multiplying the NxM matrix of returns obtained in subpoint (iv) by the Mx3 matrix of eigenvectors obtained in sub-point (viii). (x) calculating the matrix of returns to be used in the simulation by multiplying the results in sub-point (ix) with the transpose of the matrix of eigenvectors obtained in sub-point (viii). This is the set of values to be used in the simulation. (b) The curve simulation is performed as follows. (i) The time step in the simulation is one period. For each observation period in the recommended holding period select a row at random row from the calculated matrix of returns. The return for each tenor point, T, is the sum over the selected rows of the column corresponding to tenor point, T. (ii) The simulated rate for each tenor point T, is the current rate at tenor point T:

multiplied by the exponential of the simulated return
adjusted for any shifts used to ensure positive values for all tenor
point
adjusted so that the expected mean matches current expectations for
the rate at tenor point T, at the end of the recommended holding period.

For PRIIPS in Category 3 that are characterized by an unconditional protection of
capital, the PRIIP manufacturer may assume that the VaR with a 97,5% confidence level is equal to the level of the unconditional capital protection at the recommended holding period, discounted to the present date using the expected risk-free discount factor. MRM class determination for Category 4 PRIIPs
Where the PRIIP performance depends on a factor or factors unobserved in the market
or to some extent under the control of the PRIIP manufacturer, or this is the case for a component of the PRIIP, the PRIIP manufacturer shall follow the method in this
section to account for this factor or factors.

35
26. The different components of the PRIIP that contribute to the performance of the
PRIIP shall be identified, in order for those components that are not wholly or partly dependent on a factor or factors that are unobserved in the market to be treated according to the relevant methods set out in this Annex for Category 1, 2 or 3 PRIIPs. For each of these components a VaR-equivalent volatility shall be calculated.
27. The component of the PRIIP that depend wholly or partly on a factor or factors that
are unobserved in the market shall follow robust and well recognised industry and regulatory standards for determining relevant expectations as to the future contribution of these factors and the uncertainty that may exist in respect of this contribution. Where the component is not wholly dependent on a factor that is unobserved in the market, a bootstrap methodology shall be used to account for the market factors, as set out for Category 3 PRIIPs. The VaR-equivalent volatility for the component of the PRIIP shall thereby be estimated.
28. The VaR-equivalent volatility of each component of the PRIIP shall be weighted
proportionally in order to arrive at an overall VaR-equivalent volatility of the PRIIP. When weighing the components, product features shall be taken into account. Where relevant, product algorithms mitigating the market risk as well as specificities of the with-profit component shall be considered.
29. For PRIIPS in Category 4 that are characterized by an unconditional protection of
capital, the PRIIP manufacturer may assume that VaR with a 97,5% confidence level is equal to the level of unconditional capital protection at the recommended holding period, discounted to the present date using the expected risk-free discount factor.
PART 2
Methodology for assessing credit risk
I. General requirements
30. A PRIIP or its underlying investments or exposures shall be taken to entail credit
risk where the return of the PRIIP or its underlying investments or exposures depends on the creditworthiness of a manufacturer or party bound to make, directly or indirectly, relevant payments to the investor. A PRIIP with a MRM of 7 shall not need to assess credit risk.
31. Where an entity directly engages to make a payment to retail investor for a
PRIIP, credit risk shall be assessed for the entity that is the direct obligor.
32. If all payment obligations of an obligor or one or more indirect obligors are
unconditionally and irrevocably guaranteed by another entity (the guarantor), the

36 credit risk assessment of the guarantor can be used if it is more favourable than the credit risk assessment of the respective obligor or obligors.
33. Where a PRIIP invests in or is exposed to underlying investments or techniques,
including PRIIPs which themselves entail credit risk or in turn make underlying investments that entail credit risk, the credit risk shall be assessed in relation to the credit risk entailed both by the PRIIP itself and the underlying investments or exposures (including exposures to other PRIIPs), on a look-through basis and adopting a cascade assessment as necessary.
34. Where the credit risk is entailed solely at the level of underlying investments or
exposures (including to other PRIIPs), the credit risk shall not be assessed at the level of the PRIIP itself but instead at the level of these underlying investments or exposures on a look-through basis. Where the PRIIP is a UCITS or AIF, the UCITS or AIF itself shall be taken to entail no credit risk, whereas the underlying investments or exposures of the UCITS or AIF shall be assessed as necessary.
35. Where a PRIIP is exposed to multiple underlying investments entailing a credit
risk exposure, the credit risk entailed by each underlying investment representing an exposure of 10% or more of the total assets or value of the PRIIP shall be separately assessed.
36. Underlying investments or exposures to exchange-traded derivatives or cleared
OTC derivatives shall be assumed for the purposes of the credit risk assessment to carry no credit risk. No credit risk shall be taken to be entailed where an exposure is fully and appropriately collateralised, or where uncollateralised exposures that entail credit risk amount to less than 10% of the total assets or value of the PRIIP.
II. Credit risk assessment
Credit assessment of obligors
37. Where available, a PRIIP manufacturer shall define ex-ante one or more external
credit assessment institutions (ECAI) certified or registered with ESMA whose credit assessments will consistently be referred to for the purpose of the credit risk assessment. Where multiple credit assessments are available according to that policy, the median rating shall be used, defaulting to the lower of the two middle values for an even number of assessments.
38. The level of credit risk of the PRIIP and each relevant obligor shall be assessed
on the basis of, as applicable:
(a) the credit assessment assigned to the PRIIP by an ECAI; (b) the credit assessment assigned to the relevant obligor by an ECAI;

37
(c) or, in the absence of a credit assessment under either (a) or (b) or both, a default credit assessment as set out in point 43. Allocation of credit assessments to credit quality steps
39. The allocation of credit assessments of ECAIs to an objective scale of credit
quality steps shall be based on the allocation set out in Commission Implementing Regulation (EU) No …/… laying down implementing technical standards with regard to the allocation of external credit assessments of external credit assessment institutions to an objective scale of credit quality steps in accordance with Directive 2009/138/EC.8
40. In the case of credit risks assessed on a look-through basis, the credit quality step
assigned shall correspond to the weighted average credit quality steps of each relevant obligor for which a credit assessment needs to be undertaken, in proportion of the total assets they respectively represent.
41. In the case of credit risks assessed on a cascade basis, all credit risk exposures
shall be separately assessed, per layer, and the credit quality step assigned shall be the highest credit quality step, being understood that between a credit quality step set at 1 and a credit quality step set at 3, the higher of the two is 3.
42. The credit quality step pursuant to point 38, including pursuant to point 43, shall
be adjusted according to the maturity or RHP of the PRIIP, according to the following table, except where a credit assessment has been assigned that reflects the maturity or RHP of the PRIIP:
Credit quality step pursuant to point 38
Adjusted credit quality step, in the case where the maturity of the PRIIP, or its RHP when a PRIIP does not have a maturity, is up to one year Adjusted credit quality step, in the case where the maturity of the PRIIP, or its RHP when a PRIIP does not have a maturity, ranges from one year up to twelve years Adjusted credit quality step, in the case where the maturity of the PRIIP, or its RHP when a PRIIP does not have a maturity, exceeds twelve years 0 1 2 1 2 2

8
Not yet published in the Official Journal.

38
3 2 3 3
4 3 4 5
5 4 5 6
43. If the obligor has no external credit assessments, the default credit assessment as
referred to in point 38 shall be:
(a) credit quality step 3, if the obligor is regulated as a credit institution or an insurance undertaking under the applicable European regulation or regulation deemed equivalent under Union law and if the Member State where the obligor is domiciled would be allocated according to these articles to credit quality step 3; (b) credit quality step 5, for any other obligor.
III. Credit risk measure
44. A PRIIP shall be allocated to a credit risk measure (CRM) on a scale ranging
from 1 to 6, on the basis of the mapping table laid down in point 45, and by applying the credit risk mitigating factors under points 46, 47, 48 and 49, or the credit risk escalating factors under points 50 and 51, as appropriate.
45. Table on the mapping of credit quality steps into a CRM is as follows:
credit quality steps credit risk measure
0 1
2 2
3 3
4 4
5 5
6 6
46. The CRM may be assigned as 1 where the assets of a PRIIP or appropriate
collateral, or assets backing the payment obligation of the PRIIP, are:
(a) at all times until maturity equivalent to the payment obligations of the PRIIP to its investors; (b) held with a third party on a segregated account under equivalent terms and conditions as those laid down in Directive 2011/61/EU or Directive 2014/91/EU; and

39
(c) not, under any circumstances, accessible to any other creditors of the manufacturer under applicable law.
47. The CRM may be assigned as 2 where the assets of a PRIIP or appropriate
collateral, or assets backing the payment obligation of the PRIIP, are:
(a) at all times until maturity equivalent to the payment obligations of the PRIIP to its investors; (b) identified and held on accounts or registers, based on applicable law, including pursuant to Articles 275 and 276 of Solvency II; and (c) held such that the claims of retail investors have priority over the claims of other creditors of the PRIIP manufacturer or party bound to make, directly or indirectly, relevant payments to the investor.
48. Where credit risk is to be assessed on a look-through or cascade basis, the
mitigation factors under point 46 and 47 may also be applied when assessing credit risk in relation to each underlying obligor.
49. Where a PRIIP is not able to satisfy the criteria under point 47, the CRM
pursuant to point 45 may be reduced by one class where the claims of retail investors have priority over the claims of ordinary creditors, as set out in Article 108 of Directive 2014/59/EU, of the PRIIP manufacturer or party bound to make, directly or indirectly, relevant payments to the investor, in so far as the obligor is subject to relevant prudential requirements in respect of ensuring an appropriate matching of assets and liabilities.
50. The CRM pursuant to point 45 shall be increased by two classes where the claim
of a retail investor is subordinated to the claims of senior creditors.
51. The CRM pursuant to point 45 shall be increased by three classes where a PRIIP
is part of the own funds of the PRIIP obligor, as defined in article 4 (1) (118) of Regulation 575/2013 or article 93 of Directive 2009/138/EU.
PART 3
Aggregation of market and credit risk into the summary risk indicator
52. The overall summary risk indicator (SRI) is assigned according to the combination of
the CRM and the MRM classes, in accordance with the table below.
MRM class
CRM class MR
1
MR
2
MR
3
MR
4
MR
5
MR
6
MR
7
CR1 1 2 3 4 5 6 7
CR2 1 2 3 4 5 6 7
CR3 4 5 6 7

40
Monitoring data with relevance for the summary risk indicator
53. The PRIIP manufacturer shall monitor market data relevant to the calculation of the
MRM class and, if the MRM class changes to a different MRM class, the PRIIP shall attribute the corresponding MRM class to the MRM class which the PRIIP has matched for the majority of the reference points over the preceding four months.
54. The PRIIP manufacturer shall also monitor credit risk criteria relevant to the
calculation of the CRM and, if the CRM changes to a different CRM class, the PRIIP shall attribute the CRM to the relevant CRM class.
55. A review of the MRM class shall always be carried out following a decision by the
PRIIP manufacturer in respect of the PRIIP´s investment policy and/or strategy. In those circumstances, any changes to the MRM shall be intended as new determination of the PRIIP´s MRM class, and consequently, be carried out according to the general rules concerning the determination of an MRM class for the PRIIP category.
PART 4
Liquidity risk
56. A PRIIP shall be considered as having a materially relevant liquidity risk according to
the following criteria:
(a) the PRIIP is admitted to trading on a secondary market or alternative liquidity facility and there is no committed liquidity offered by market makers and/or the PRIIP manufacturer, so that the liquidity depends only on the availability of buyers and sellers on the secondary market or alternative liquidity facility, taking into account that regular trading of a product at one point in time does not guarantee the regular trading of the same product at any other point in time; or (b) the average liquidity profile of the underlying investments is significantly lower than the regular reimbursement frequency for the PRIIP, when and to the extent liquidity offered by the PRIIP is conditional to the liquidation of its underlying assets. (c) the PRIIP manufacturer estimates that the retail investor may face significant difficulties in terms of time and/or costs for disinvesting during the life of the product, subject to specific market conditions. CR4 5 6 7 CR5 6 7 CR6 7

41
57. A PRIIP shall be considered illiquid, whether contractually or not, according to the
following criteria:
(a) the PRIIP is not admitted to trading on a secondary market, and no alternative liquidity facility is promoted by the PRIIP manufacturer or a third party, or the alternative liquidity facility is subject to significant limiting conditions, including significant early exit penalties or discretionary redemption prices, or where there is an absence of liquidity arrangements; (b) the PRIIP offers potential early exit or redemption possibilities prior to the applicable maturity, but these are subject to significant limiting conditions, including significant exist penalties or discretionary redemption prices, or being subject to the prior consent and discretion of the PRIIP manufacturer; (c) the PRIIP does not offer potential early exit or redemption possibilities prior to the applicable maturity.
58. A PRIIP shall be considered liquid in all other cases.

42
ANNEX III
PRESENTATION OF SUMMARY RISK INDICATOR
Presentation format

PRIIP manufacturers shall use the format below for the presentation of the Summary
Risk Indicator (SRI) in the key information document. The relevant number shall be highlighted as shown depending on the SRI for the PRIIP. Completion guidance with regard to the SRI
The narrative explanation after the SRI briefly explains the purpose of the SRI and the
underlying risks.
Immediately below the indicator, the time frame of the recommended holding period is
indicated. In addition, a warning shall be included directly below the indicator; (a) When the risk of the PRIIP is considered to be significantly higher if the holding period is different. (b) When a PRIIP is considered to have a materially relevant liquidity risk or to be illiquid, depending on whether this is contractual in nature or not.
As applicable for each PRIIP, the narrative explanation includes:
(a) a warning in bold font where:
(i) a PRIIP is considered to have currency risk under Part 6 of Annex II (Element C); (ii) a clarification that the PRIIP holds a possible obligation to add to the initial investment (Element D)

43
(b) Where applicable, the narrative explanation below needs to explain risks materially relevant to the PRIIP which could not be adequately captured by the SRI (Element E) (c) A clarification:
(i) that the PRIIP holds a (partial) capital protection against market risk where relevant, including a specification of the percentage of the invested capital that is protected (Element F); (ii) of the specific conditions of the limitations where the (partial) capital protection against market risk is limited (Element G); (iii) that the PRIIP holds no capital protection against market risk, where relevant (Element H) (iv) that the PRIIP holds no capital guarantee against credit risk, where relevant (Element I); and (v) of the specific conditions of the limitations where the protection against credit risk is limited (Element J).
5. For exchange-traded derivatives that are futures, call options and put options traded on
a regulated market or on a third-country market considered to be equivalent to a regulated market in accordance with Article 28 of Regulation (EU) 600/2014, and as such do not fall within the definition of an OTC derivative as defined in Article 2(7) of Regulation (EU) 648/2012, Elements A, B, and, where relevant, H, shall be included. Narrative explanations
6. For the purposes of the SRI presentation, including point 4, the following narrative
explanations shall be used, as appropriate:
[Element A] The summary risk indicator is a guide to the level of risk of this product compared to other products. It shows how likely it is that the product will lose money because of movements in the markets or because we are not able to pay you. [Element B] We have classified this product as [1/2/3/4/5/6/7] out of 7, which is [1=”the lowest” / 2=“a low” / 3=“a medium-low” / 4=“a medium” / 5=“a medium-high” / 6=“the second-highest” / 7=”the highest“] risk class. [In addition, insert a brief explanation of the classification of the product with a maximum of 300 characters in plain language] [An example explanation: This rates the potential losses from future performance at a [1=”very low”/2=”low”/3=”medium-low”/4=”medium”/5=”mediumhigh”/6=”high”/7=”very high”] level, and poor market conditions [1, 2= “are very unlikely to”/3=”are unlikely to”/4=“could”/5=“will likely”/6=”are very likely to”] impact [our] [the] capacity [of X] to pay you].

44
[[Where applicable:] Element C, in bold] Be aware of currency risk. You will receive payments in a different currency, so the final return you will get depend on the exchange rate between the two currencies. This risk is not considered in the indicator shown above. [[Where applicable:] Element D] In some circumstances you may be required to make further payments to pay for losses. (in bold) The total loss you may incur may significantly exceed the amount invested. [Where applicable:] [Element E] [Other risks materially relevant to the PRIIP not included in the summary risk indicator to be explained with a maximum of 200 characters] [Where applicable:] [Element F] [You are entitled to receive back at least [insert %] of your capital. Any amount over this, and any additional return, depends on future market performance and is uncertain.] [Where applicable:] [Element G] [However, this protection against future market performance will not apply if you [..]

[Where early exit conditions apply] cash-in before [… years/months/days]]
[Where ongoing payments must be made] fail to make your payments in time.
[Where other limitations apply: explain these in a maximum of […] characters in
plain language.]
[Where applicable:] [Element H] [This product does not include any protection from future market performance so you could lose some or all of your investment.] [Where applicable:] [Element I] [If (we) (are) not able to pay you what is owed, you could lose your entire investment.] [Where applicable:] [Element J] [However, you may benefit from a consumer protection scheme (see the section “what happens if we are unable to pay you”). The indicator shown above does not consider this protection.]

45
ANNEX IV
PERFORMANCE SCENARIOS
Number of scenarios

The three performance scenarios under this Regulation, are to show a range of
possible returns and shall be as follows:
(a) An unfavourable scenario;
(b) A moderate scenario; and
(c) A favourable scenario.
An additional scenario shall set out significant unfavourable impacts of the product
where these would not be adequately covered in the unfavourable scenario in accordance with point 1. The additional scenario shall show intermediate periods where these would be shown for the performance scenarios under point 1.
An additional scenario for insurance-based investment products shall be based on
the moderate scenario in accordance with point 1(b) above, where the performance is relevant in respect of the return of the investment. Calculation of scenario values for the recommended holding period
The scenario values under different performance scenarios shall be calculated in a
similar manner to the market risk measure. The scenarios shall be calculated for the recommended holding period.
The unfavourable scenario shall be the value of the PRIIP at the 10 th percentile.
The moderate scenario shall be the value of the PRIIP at the 50 th percentile.
The favourable scenario shall be the value of the PRIIP value at the 90 th percentile.
For Category 2 PRIIPs, the expected values at the recommended holding period
shall be:
(a) Unfavourable:
Exp [ M1*N +  √N * (-1.28 + 0.107 * µ1 / √N + 0.0724 * µ2 / N - 0.0611 * µ1 2 / N) – 0.5 2N ] (b) Moderate:

46
Exp [ M1N -  µ1/6 – 0.5
2N ]
(c) Favourable:
Exp [ M1N +  √N * ( 1.28 + 0.107 * µ1 / √N – 0.0724 * µ2 / N + 0.0611 * µ1 2 / N) – 0.5 2N ] where N is the number of periods in the recommended holding period, and where the other terms are as defined in Annex II point 12.
9. For Category 3 PRIIPs, the following adjustments shall be made to the calculation
of performance scenarios:
(a) the expected return for each asset or assets shall be the return observed over the period as determined under point 6 of Annex II Part 1; and (b) the expected performance shall be calculated at the end of the recommended holding period, and without discounting the expected performance using the expected risk-free discount factor.
10. For Category 4 PRIIPs, the method under point 26 of Annex II shall be used in
respect of those factors that are not observed in the market, combined as necessary with the method for Category 3 PRIIPs. The relevant methods for Category 2 and 3 PRIIPs set out in points 8 and 9 shall be used for the relevant components of the PRIIP where the PRIIP combines different components. The performance scenarios shall be a weighted average of the relevant components. Product features and capital guarantees shall be taken into consideration in the performance calculations.
11. Category 1 PRIIPs as defined by Annex II Part 1 point 4(a), and Category 1 PRIIPs
as defined in Annex II part 1 point 4(b) that are not exchange-traded derivatives that are futures, call options and put options traded on a regulated market or on a third-country market considered to be equivalent to a regulated market in accordance with Article 28 of Regulation (EU) 600/2014, and as such do not fall within the definition of an OTC derivative as defined in Article 2(7) of Regulation (EU) 648/2012, shall calculate performance scenarios in accordance with point 9.
12. Category 1 PRIIPs, that are exchange-traded derivatives that are futures, call
options and put options traded on a regulated market or on a third-country market considered to be equivalent to a regulated market in accordance with Article 28 of Regulation (EU) 600/2014, and as such do not fall within the definition of an OTC derivative as defined in Article 2(7) of Regulation (EU) 648/2012, shall show performance scenarios in the form of pay-off structure graphs. A graph in accordance with Template C of Annex V will be included to show performance for all scenarios for the different levels of the underlying value. The horizontal axis of

47 the graph shows the various possible prices of the underlying value and the vertical axis shows the profit or loss at the different prices of the underlying value. For every price of the underlying value, the graph shall show the resulting profit or loss and at which price of the underlying value the profit or loss shall be zero.
13. Category 1 PRIIPs as defined by Annex II Part 1 point 4(c) shall provide a
reasonable and conservative best estimate of the expected values for the performance scenarios at the recommended holding period. (a) The scenarios selected and shown shall be consistent with and complement the other information contained in the key information document, including the overall risk profile for the PRIIP. The PRIIP manufacturer shall ensure the consistency of the scenarios with internal product governance conclusions, including amongst other things, any stress-testing undertaken by the PRIIP manufacturer for the PRIIP, and data and analysis used for the purposes of producing the other information contained with the key information document. (b) The scenarios shall be selected to give a balanced presentation of the possible outcomes of the product in both favourable and unfavourable conditions, but only scenarios that can be reasonably expected shall be shown. The scenarios shall not be selected so as give undue prominence to favourable outcomes at the expense of unfavourable ones. Calculation of expected values for intermediate holding periods
14. For PRIIPs with a recommended holding period between 1 and 3 years,
performance shall be shown at 2 different holding periods: 1 year and at the end of the recommended holding period.
15. For PRIIPs with a recommended holding period of 3 years or more, performance
shall be shown at 3 holding periods: 1 year, half the recommended holding period rounded up to the nearest year, and the recommended holding period.
16. For PRIIPs with a recommended holding period of 1 year or less, no performance
scenarios for intermediate holding periods shall be shown.
17. For Category 2 PRIIPs, the values to be shown for the intermediate periods, shall
be calculated using the formulas in point 8 with the N defined to be the number of trading periods from the start date to the end of the intermediate period.
18. For Category 1 and Category 4 PRIIPs, the values to be shown for the intermediate
periods shall be estimated by the PRIIP manufacturer in a manner consistent with the estimation at the end of the recommended holding period. To this end, the method used to estimate the value of the PRIIP at the start of each intermediate

48 period needs to produce the same value for the entire recommended holding period, as under the method prescribed in points 11 and 10 respectively.
19. For category 3 PRIIPs, to produce the favourable, moderate and unfavourable
scenarios at an intermediate period before the RHP, the manufacturer shall pick three underlying simulations as referred to in points 16 to 24 of Annex II used for the calculation of the MRI, on the basis of underlying levels only and such that the simulated value of the PRIIPS on such period is likely to be consistent with the relevant scenario. (a) To produce the favourable, moderate and unfavourable scenarios at an intermediate period for a category 3 PRIIP with one underlying and whose value is known to be an increasing function of its underlying level, the manufacturer shall pick three underlying simulations as referred to in points 16 to 24 of Annex II used for the calculation of the MRI, leading respectively to the 90% for the favourable scenario, 50% for the moderate scenario and 10% for the unfavourable scenario percentile levels of the underlying.. (b) To produce the favourable, moderate and unfavourable scenarios at an intermediate period for a category 3 PRIIP with one underlying and whose value is known to be an decreasing function of its underlying level, the manufacturer shall pick three underlying simulations as referred to in points 16 to 24 of Annex II used for the calculation of the MRI, leading respectively to the 90% for the unfavourable scenario, 50% for the moderate scenario and 10% for the favourable scenario percentile levels of the underlying (c) To produce the favourable, moderate and unfavourable scenarios at an intermediate period for a category 3 PRIIP other than those mentioned in points 19 (a) and 19 (b) the manufacturer shall choose underlying values consistent with the 90%, 50% and 10% percentile levels of the PRIIP and use these values as the seed values for a simulation to determine the value of the PRIIP.
20. For Category 1 PRIIPs, that are exchange-traded derivatives that are traded on a
regulated market or on a third-country market considered to be equivalent to a regulated market in accordance with Article 28 of Regulation (EU) 600/2014, and as such do not fall within the definition of an OTC derivative as defined in Article 2(7) of Regulation (EU) 648/2012, performance scenarios for intermediate holding periods shall not be included.
21. For all scenarios at intermediate periods, the estimate of the distribution used to
read the value of the PRIIP at different percentiles shall be consistent with the

49 observed return and volatility observed over the past 5 years of all market instruments that determine the PRIIP’s value.
22. The unfavourable scenario shall be the estimate of the value of the PRIIP at the
start of the intermediate period consistent with the 10th percentile.
23. The moderate scenario shall be the estimate of the value of the PRIIP at the start of
the intermediate period consistent with the 50th percentile.
24. The favourable scenario shall be the estimate of the value of the PRIIP at the start
of the intermediate period consistent with the 90th percentile.
General requirements
25. The performance of the PRIIP shall be calculated net of all applicable costs in
accordance with Annex VI for the scenario and holding period being presented.
26. Performance shall be presented in monetary units. The amounts used shall be
consistent with those under point 91 in Annex VI.
27. Performance shall also be presented in percentage terms, as the average annual
return of the investment. That figure shall be calculated considering net performance as numerator and the initial investment amount or the price paid as denominator. For those PRIIPs where there is no initial investment or price paid such as future contracts or swaps, the percentage shall be calculated considering the nominal value of the contract and a foot note added to explain it.
28. For an insurance based investment product the following shall apply in addition
when calculating the performance scenarios in respect of the investment:
(a) Future profit participation shall be taken into account.
(b) Assumptions on future profit participation shall be consistent with the assumption on the annual rates of return of the underlying assets. (c) Assumptions on how future profits are shared between the PRIIP manufacturer and the retail investor and other assumptions on future profit sharing shall be realistic and in line with the current business practice and business strategy of the PRIIP manufacturer. Where there is sufficient evidence that the undertaking will change its practices or strategy, the assumptions on future profit sharing shall be consistent with the changed practices or strategy. For life insurers within the scope of Directive 2009/138/EC of the European Parliament and the Council 9 ,

9
Directive 2009/138/EC of the European Parliament and of the Council of 25 November 2009 on the taking-up and pursuit of the business of Insurance and Reinsurance (Solvency II), OJ L 335, 17.12.2009, p. 1.

50 these assumptions shall be consistent with the assumptions on future management actions used for the valuation of technical provisions in the Solvency II-balance-sheet. (d) Where a component of the performance relates to profit participation that is payable on a discretionary basis, this component shall only be assumed in the favourable performance scenarios. (e) The performance scenarios shall be calculated on the basis of the investment amounts set out in point 26.

51
ANNEX V
METHODOLOGY FOR THE PRESENTATION OF PERFORMANCE SCENARIOS
PART 1
General presentation specifications

The performance scenarios shall be presented in a way that is fair, accurate, clear and
not misleading, and that is likely to be understood by the average retail investor.
Where under Annex IV point 16 performance scenarios may be shown only at maturity
or at the recommended holding period, this shall be clearly explained in the narrative set out in element D in Part 2 of this Annex.
In all cases, the narrative explanations set out in elements A, B, C and E in Part 2 of
this Annex shall be included, except in the case of Category 1 PRIIPs according to
Annex IV point 12, where the narrative explanations set in elements F to K shall be
used instead.
PART 2
Presentation of Performance Scenarios
PRIIP manufacturers shall present the performance scenarios by means of the format below. Dependent on the recommended holding period the interim period differs. For regular premium insurance-based investment products an additional row containing the cumulative biometric risk premium is included before the row of “What you might get back after costs” in Template B. Returns for this scenario shall only be shown in absolute values. For Category 1 PRIIPs according to Annex IV point 12, the performance scenarios shall be displayed according to Template C. Template A: Single investment and/or single premium paid.

52
Template B: regular investments and/or premiums are paid.
Template C: exchange-traded derivatives.

53
Performance scenarios
[Element A] This [table/graph] shows the money you could get back over the next [recommended holding period] years, under different scenarios, assuming that you invest €[…] [per year]. [Element B] The scenarios shown illustrate how your investment could perform. You can compare them with the scenarios of other products. [Element C]The scenarios presented are an estimate of future performance based on evidence from the past, and are not an exact indicator. What you get will vary depending on how the market performs and how long you keep the investment/product. [] [Where applicable][Element D] This product cannot be [easily] cashed in. This means it is difficult to estimate how much you would get back if you cash in before [the recommended holding period/maturity]. You will either be unable to cash in early or you will have to pay high costs or make a large loss if you do so. [Element E] The figures shown include all the costs of the product itself, [where applicable]:[but may not include all the costs that you pay to your advisor or distributor][and includes the costs of your advisor or distributor]. The figures do not take into account your personal tax situation, which may also affect how much you get back. [Element F] This graph illustrates how your investment could perform. You can compare them with the pay-off graphs of other derivatives. [Element G] The graph presented gives a range of possible outcomes and is not an exact indication of what you might get back. What you get will vary depending on how the underlying will develop. For each value of the underlying, the graph shows what the profit or loss of the product would be. The horizontal axis shows the various possible prices of the underlying value on the expiry date and the vertical axis shows the profit or loss. [Element H] Buying this product holds that you think the underlying price will [increase/decrease].

54
[Element I] Your maximum loss would be that you will lose all your investment (premium paid). [Element J] The figures shown include all the costs of the product itself, but may not include all the costs that you pay to your advisor or distributor. The figures do not take into account your personal tax situation, which may also affect how much you get back.

55
ANNEX VI
METHODOLOGY FOR THE CALCULATION OF COSTS
PART 1
List of costs
I. List of Costs of investments funds (AIFs and UCITS)
Costs to be disclosed
One-off costs

A one-off cost is an entry or exit cost which is either:
a) paid directly by the retail investor; or b) deducted from a payment received by or due to the retail investor.
One-off costs are costs borne by the retail investor that are not deducted from the assets
of the AIF or UCITS.
One-off costs include, but are not limited to, the following types of up-front initial
costs that shall be taken into account in the cost amount to be disclosed in the key information document:
(a) Distribution fee, to the extent that the amount is known to the management company. If the actual amount is not known to the management company, the maximum of the possible known distribution costs for the specific PRIIP shall be shown; (b) Constitution costs (up-front part); (c) Marketing costs (up-front part); (d) Subscription fee including taxes. Recurring Costs
Recurring costs are payments deducted from the assets of an AIF or UCITS, and
represent:
a) expenses necessarily incurred in their operations; b) any payments, including remunerations, to parties connected with the AIF or UCITS or providing services to them; or c) transaction costs.
Recurring costs include, but are not limited to, the following types of costs that are
deducted from the assets of the AIF or UCITS, and shall be taken into account in the cost amount to be disclosed in the key information document:
(a) all payments to the following persons, including any person to whom they have delegated any function:
(i) the management company of the fund,
(ii) directors of the fund if an investment company, (iii)the depositary,

56
(iv)the custodian(s),
(v) any investment adviser;
(b) all payments to any person providing outsourced services to any of the above, including:
(i) providers of valuation and fund accounting services, (ii) shareholder service providers, such as the transfer agent and broker dealers that are record owners of the fund’ shares and provide sub-accounting services to the beneficial owners of those shares, (iii) providers of collateral management services, (iv)providers of prime-brokerage services, (v) securities lending agents, (vi) providers of property management and similar services; (c) registration charges, listing fees, regulatory charges and similar charges, including passporting fees; (d) provisioned fees for specific treatment of gain and losses; (e) audit fees; (f) payments to legal and professional advisers; (g) any costs of distribution or marketing, to the extent that the amount is known to the management company. If the actual amount is not known to the management company, the maximum of the possible known distribution costs for the specific PRIIP shall be shown; (h) financing costs, related to borrowing (provided by related parties); (i) costs of capital guarantee provided by a third party guarantor; (j) payments to third parties to meet costs necessarily incurred in connection with the acquisition or disposal of any asset in the fund’s portfolio (including transaction costs as referred to in points 7-23); (k) the value of goods or services received by the management company or any connected person in exchange for placing of dealing orders; (l) where a fund invests its assets in UCITS or AIFs, its summary cost indicator shall take account of the charges incurred in the UCITS/AIFs. The following shall be included in the calculation:
(i) if the underlying is a UCITS or AIF its most recently available summary cost indicator figure shall be used; this may be the figure published by the UCITS or AIF or its operator or management company, or a figure calculated by a reliable third-party source if more up-to-date than the published figure; (ii) the summary cost indicator may be reduced to the extent that there is any arrangement in place (and that is not already reflected in the fund’s profit and

57 loss account) for the investing fund to receive a rebate or retrocession of charges from the underlying AIF/UCITS; (iii) where the acquisition or disposal of units does not occur at the mid price of the UCITS/AIF, the value of the difference between the transaction price and the mid price shall be taken into account as transaction costs, to the extent that this is not included in the summary cost indicator. (m)where a fund invests in a PRIIP other than UCITS or AIFs, its summary cost indicator shall take account of the charges incurred in the underlying PRIIP. The following shall be included in the calculation:
(i) the most recently available summary cost indicator of the underlying PRIIP shall be included in the calculation; (ii) the summary cost indicator may be reduced to the extent that there is any arrangement in place (and that is not already reflected in the fund’s profit and loss account) for the investing fund to receive a rebate or retrocession of charges from the underlying PRIIP; (iii) in cases where the acquisition or disposal of units does not occur at the mid price of the underlying PRIIP, the value of the difference between the transaction price and the mid price shall be taken into account as transaction costs, to the extent that this is not included in the summary cost indicator. (n) where a fund invests in an investment product other than a PRIIP its summary cost indicator shall take account of the charges incurred in the underlying investment product. The PRIIP manufacturer shall either use any published information that represents a reasonable substitute for summary cost indicator or else shall make a best estimate of its maximum level based on scrutiny of the investment product’s current prospectus and most recently published report and accounts; (o) operating costs (or any remuneration) under a fee-sharing arrangement with a third party to the extent that they have not been already included in another type of cost mentioned above; (p) earnings from efficient portfolio management techniques if they are not paid into the portfolio; (q) implicit costs incurred by structured funds as referred to in section II, and notably points 37 to 47; (r) dividends served by the shares held in the portfolio of the funds, shall the dividends not accrue to the fund. Incidental costs
6. The following list indicates the types of incidental costs that shall be taken into account
in the amount to be disclosed:
(a) a performance–related fee payable to the management company or any investment adviser, including performance fees as referred to in point 24; (b) carried interests as referred to in point 25.

58
Calculation of specific types of costs of investments funds Transaction costs
7. Transaction costs shall be calculated on an annualised basis, based on an average of the
transaction costs incurred by the PRIIP over the previous three years. Where the PRIIP has been operating for less than three years, transaction costs shall be calculated using the methodology set out in point 21.
8. The aggregate transaction costs for a PRIIP shall be calculated as the sum of the
transaction costs (calculated as set out below) in the base currency of the PRIIP for all individual transactions undertaken by the PRIIP in the specified period. This sum shall be converted into a percentage by dividing by the average net assets of the PRIIP over the same period.
9. When calculating the transaction costs incurred by the PRIIP over the previous three
years, actual transaction costs must be calculated using the methodology described below in points 12 to 18 for investments in the following instruments:
(a) transferable securities as defined by Article 2 of Directive 2007/16/EC of 19 March 2007 implementing Council Directive 85/611/EEC on the coordination of laws, regulations and administrative provisions relating to undertakings for collective investment in transferable securities (UCITS) as regards the clarification of certain definitions; (b) other instruments that there are frequent opportunities to dispose of, redeem, or otherwise realise at prices that are publicly available to market participants and that are either market prices or prices made available, or validated, by valuation systems independent of the issuer.
10. Estimates of transaction costs using the methodology described below in points 19 to
20 must be used for investments in other instruments or assets.
Treatment of anti-dilution mechanisms
11. Where a PRIIP has a pricing mechanism that offsets the impact of dilution from
transactions in the PRIIP itself, the amount of benefit accruing to the ongoing holders of the PRIIP from anti-dilution mechanisms may be deducted from the transaction costs incurred within the PRIIP using the following methodology:
(a) The monetary amount of any anti-dilution levy, or other payment in connection with a transaction in the PRIIP itself, that is paid to the PRIIP may be subtracted from the total transaction costs (b) The benefit to the PRIIP of issuing units (or otherwise enabling investment in the PRIIP) at a price other than the mid price, or of cancelling units (or otherwise enabling redemption of funds from the PRIIP) at a price other than the mid price, provided that the PRIIP itself receives the benefit, shall be calculated as follows and may be subtracted from the total transaction costs:
(i) The difference between the price of units issued and the mid price, multiplied by the net number of units issued;

59
(ii) The difference between the price of units cancelled and the mid-price, multiplied by the net number of units cancelled. Actual transaction costs
12. The actual transaction costs for each transaction shall be calculated on the following
basis:
(a) for each purchase undertaken by the PRIIP, the price of the instrument at the time the purchase order is transmitted to another person for execution (the ‘arrival price’) shall be subtracted from the net realised execution price of the transaction. The resulting value shall be multiplied by the number of units purchased; (b) for each sale undertaken by the PRIIP, the net realised execution price of the transaction shall be subtracted from the price of the instrument at the time the order to sell is transmitted to another person for execution (the ‘arrival price’). The resulting value shall be multiplied by the number of units sold.
13. The net realised execution price shall be calculated as the price at which the transaction
is executed, including all charges, commissions, taxes and other payments (for example anti-dilution levies) associated with the transaction, either directly or indirectly, where those payments are made from the assets of the PRIIP.
14. The arrival price shall be calculated as the mid-market price of the investment at the
time when the order to transact is transmitted to another person. For orders that are transacted on any day that is not the day that the order was originally transmitted to another person, the arrival price shall be calculated as the opening price of the investment on the day of the transaction or, where the opening price is not available, the previous closing price. Where a price is not available at the time when the order to transact is transmitted to another person (for example because the order has been initiated outside market opening hours or in over-the-counter markets where there is no transparency of intra-day prices), the arrival price shall be calculated as the opening price on the day of the transaction or, where the opening price is not available, the previous closing price. Where an order is executed without being transmitted to another person, the arrival price shall be calculated as the mid-market price of the investment at the time when the transaction was executed.
15. Where information about the time when the order to transact is transmitted to another
person is not available (or not available to a sufficient level of accuracy), or where information about the price at that time is not available, it is permissible to use as the arrival price the opening price of the investment on the day of the transaction or, where the opening price is not available, the previous closing price. When calculating transaction costs using data prior to 31 December 2016, it may be considered that intraday prices are not available.
16. Costs associated with transactions in financial instruments, undertaken by PRIIPs, that
fall within one of the categories referred to in items 4 to 10 of Section C of Annex I to Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU Text with EEA relevance shall be calculated in the following way:

60
(a) For instruments that are standardised and where there is regular trading in the instrument itself (for example an index future on a major equity index), transaction costs shall be calculated with reference to the instrument itself. The arrival price shall be calculated as the mid price of the instrument; (b) For linear instruments that are customised, and where there is no price transparency or regular trading in the instrument itself, transaction costs shall be calculated with reference to the underlying asset(s). The arrival price shall be calculated based on the price(s) of the underlying assets, using appropriate weightings if there is more than one underlying asset. Where the cost of transacting in the instrument is materially higher than the cost of transacting in the underlying asset, this must be reflected in the transaction cost calculation; (c) For non-linear instruments, it is permissible to calculate the transaction costs as the difference between the price paid or received for the instruments and the fair value of the instrument, on the basis described in points 37-47 below.
17. In calculating the costs associated with foreign exchange, the arrival price must reflect
a reasonable estimate of the consolidated price, and must not simply be the price available from a single counterparty or foreign exchange platform, even if an agreement exists to undertake all foreign exchange transactions with a single counterparty.
18. In calculating the costs associated with orders that are initially entered into an auction,
the arrival price shall be calculated as the mid price immediately prior to the auction. Transaction costs for other assets
19. When estimating transaction costs for assets other than assets in point 9, the
methodology in point 12 should be used with the definition of arrival price as follows:
(a) for a sale:
(i) the arrival price shall be calculated as the previous independent valuation price of the asset, adjusted for market movements, where appropriate, using an appropriate benchmark index, (ii) where a previous independent valuation price is not available, the transaction costs must be estimated based on the difference between the transaction price and an appraisal of the fair value of the asset prior to sale; (b) for a purchase:
(i) the arrival price shall be calculated as the previous independent valuation price of the asset, adjusted for market movements, where appropriate, using an appropriate benchmark index, where such a price is available, (ii) where a previous independent valuation price is not available, the transaction costs must be estimated based on the difference between the transaction price and an appraisal of the fair value of the asset prior to purchase.

61
20. The transaction cost estimate must not be less than the amount of actual identifiable
costs directly associated with the transaction.
Transaction costs for new PRIIPs
21. For PRIIPs that have been operating for less than 3 years and that invest
predominantly in assets in point 9, transaction costs may be calculated either by multiplying an estimate of portfolio turnover in each asset class according to the methodology referred to in (c), or as an average of the actual transaction costs incurred during the period of operation and a standardised estimate on the following basis:
(a)For each six months that the PRIIP has been operating, transaction costs shall be calculated on the basis described in points 12 to 18; (b)For other periods of six months, transaction costs shall be estimated by multiplying an estimate of portfolio turnover in each asset class according to the methodology referred to in (c); (c)The methodology to follow shall be that under (i), (ii) or (iii) according to the asset class:
(i) For the following asset classes, transaction costs can be calculated as the average of the estimated cost of transaction (based on bid-ask spreads divided by two) in this asset class under normal market conditions. To estimate the cost, one or more reference indexes should be identified for each asset class. Then, the average bid-ask spreads of the underlying indexes should be collected. The data collected should refer to the closing bid-ask spread at the [tenth] business day [or another business day] of the each month during the last year [or the last two years]. The bid-ask spreads collected is then divided by two to obtain the estimated cost of transaction of each point in time. The average of this values is the estimated cost of transaction in each asset class under normal market conditions. Asset Classes Sub Asset Classes Government bonds Government bonds and similar instruments developed market rating AAA-A Government bonds and similar instruments developed market different rating below A Government bonds emerging markets (hard and soft currency) Government bonds emerging markets (hard and soft currency)

62
Investment grade corporate bonds
Investment grade corporate bonds
Other corporate bonds High yield corporate bonds (ii) For the following asset classes, transaction costs (including explicit costs and implicit costs) shall be estimated either by using comparable information or by adding estimates of explicit costs to estimates of half the bid-ask spread, using the methodology described under (i). Asset Classes Sub Asset Classes Liquidity Money market instruments (for the sake of clarity, money markets funds not included) Shares developed markets Large-cap shares (developed markets) Mid-cap shares (developed markets) Small-cap shares (developed markets) Shares emerging markets Large-cap shares (emerging markets) Mid-cap shares (emerging markets) Small-cap shares (emerging markets) Listed derivatives Listed derivatives (iii) For the following asset classes the transaction cost is the average of the observed cost of transaction (based on bid-ask spreads divided by two) in this asset class under normal market conditions. In identifying the observed cost of transaction results of a panel survey may be taken into account. Asset Classes Sub Asset Classes OTC OTC Exotic options OTC Plain vanilla options OTC IRS, CDS and similar OTC Swaps and similar instruments (different from IRS, CDS and similar) OTC FX Forwards developed markets OTC FX Forwards emerging markets
22. Estimates of portfolio turnover for a PRIIP that has been operating for less than one
year must be made on a consistent basis with the investment policy disclosed in the

63 offering documents. Estimates of portfolio turnover for a PRIIP that has been operating for more than one year must be consistent with actual portfolio turnover.
23. For PRIIPs that have been operating for less than three years and that invest
predominantly in assets other than assets in point 9 the PRIIP manufacturer shall estimate the transaction costs on the basis of the fair value method using comparable assets. Performance related fees
24. To calculate performance related fees the following steps shall be taken:
(a) compute the fees on the basis of historical data covering the last 5 years. The average annual performance fees shall be computed in percentage terms, (b) where a full performance fees history is not available because the fund/share class is new or the fund’s terms have changed due to the introduction of the performance fee or the change of one of its parameters, the abovementioned method shall be adjusted according to the following steps:
(i) take the relevant available history of the performance fees of the fund/share class; (ii) for any years for which data is not available:

estimate the return of the fund/ share class,
for new funds, their return shall be estimated using the return of a comparable fund or of a peer group. The estimated return shall be gross of all the costs charged to the new fund. Therefore peer groups´ returns need to be adjusted by adding the average relevant costs charged according to the rules of the new fund. For instance, in case of a new class with a different fee structure, the returns of the existing class shall be adjusted taking into account the costs of the new class;
in case of a relative performance fee model, take into account the
historical series of the benchmark/hurdle rate.
(iii) compute the fees from the beginning of the sample period, as required in sub-point (a), until the date of availability of the actual performance fee data of the fund, applying the relevant algorithm to the abovementioned historical series; (iv) concatenate both performance fee series to one series over the full sample period as required in sub-point (a); (v) compute the performance fees according sub-point (a) methodology (average of annual performance fees). Carried interests

To calculate carried interests the following steps shall be taken:

64
(a) compute the fees on the basis of historical data covering the last 5 years. The average annual carried interests shall be computed in percentage terms; (b) where a full carried interests history is unavailable because the fund/share class is new or the fund’s terms have changed due to the introduction of carried interests or the change of one of its parameters, the abovementioned method shall be adjusted according to the following steps:
(i) take the relevant available history of the carried interests of the fund/ share class; (ii) for any years for which data is not available:

estimate the return of the fund/ share class,
for new funds, their return shall be estimated using the return of a comparable fund or of a peer group. The estimated return shall be gross of all the costs charged to the new fund. Therefore peer group’s returns need to be adjusted by adding the average relevant costs charged according to the rules of the new fund. For instance, in case of a new class with a different fee structure, the returns of the existing class shall be adjusted taking into account the costs of the new class. (iii) compute the carried interests from the beginning of the sample period, as required in sub-point (a), until the date of availability of the actual carried interests data of the fund, applying the relevant algorithm to the abovementioned historical series; (iv) concatenate both carried interests series to one series over the full sample period as required in sub-point (a); (v) compute the carried interests according to the sub-point (a) methodology (average of annual carried interests).

If no carried interests are taken throughout the investment, a warning needs to
accompany the indication of zero carried interests in the composition of costs table in order to clarify that a payment of x% of the final return shall take place subsequently to the exit of the investment.
II. List of costs of PRIPs other than investment funds
Costs to be disclosed
One-off costs
A one-off cost is an entry and exit cost which include initial charges, commissions or
any other amount paid directly by the retail investor or deducted from a payment received by or due to the retail investor.
One-off costs are borne by a PRIP other than an investment fund, whether they
represent expenses necessarily incurred in its operation, or the remuneration of any party connected with it or providing services to it. One-off entry costs and charges

65
29. One-off entry costs and charges include, but are not limited to, the following types that
shall be taken into account in the cost amount to be disclosed for PRIPs other than investment funds:
(a) sales commissions;
(b) structuring costs, including market-making costs (spread) and settlement costs; (c) hedging costs (to ensure that the PRIIP manufacturer is able to replicate the performance of the derivative component of the structured product – these costs include transaction costs) (d) legal fees; (e) costs for capital guarantee; (f) implicit premium paid to the issuer. One-off exit costs and charges
30. One-off exit costs and charges include, but are not limited to, the following types that
shall be taken into account in the amount to be disclosed for PRIPs other than investment funds:
(a) proportional fees;
(b) bid-mid spread to sell the product and any explicit costs or penalties for early exit applicable. The estimation of the bid-mid spread shall be done in relation to the availability of a secondary market, to the market conditions and the type of product. In the situation where the PRIIP manufacturer (or a related third party) is the only available counterparty to buy the product on the secondary market, it shall estimate the exit costs to be added to the fair value of the product according to its internal policies.
31. contract-for-difference (CFD) related costs such as:
(a) commissions charged by CFD providers – general commission or a commission on each trade - i.e. on opening and closing a contract; (b) CFD trading such as bid-ask spreads, daily and overnight financing costs, account management fees and taxes which are not already included in the fair value. Recurring Costs
32. Recurring costs are payments regularly deducted from all payments due to the retail
investor or from the amount invested.
33. Recurring costs include all types of cost borne by a PRIP other than an investment fund
whether they represent expenses necessarily incurred in its operation, or the remuneration of any party connected with it or providing services to it.
34. The following list is indicative but not exhaustive of the types of recurring charge that,
if they are deducted or charged separately, shall be taken into account in the amount to be disclosed:
(a) costs related to coupon payments;

66
(b) costs of the underlying, if any.
Costs of PRIPs referred to in Annex III point 12
35. One-off entry costs and charges. One-off exit costs and charges are exchange fees,
clearing fees and settlement fees if known.
36. Recurring costs. Recurring costs are hedging costs borne under normal market
conditions and stressed market conditions.
Calculation of implicit costs of PRIPs other than investment funds
37. For the purposes of the calculation of the implicit costs embedded in PRIPs, the PRIIP
manufacturer shall refer to the issue price and, after the subscription period, to the price available to purchase the product on a secondary market.
38. The difference between the price and the fair value of the product is considered as an
estimation of the total entry costs included in the price. If the PRIIP manufacturer is unable to distinguish the relevant implicit costs to be disclosed as referred to in point 29 using the aforementioned difference between the price and the fair value, it shall liaise with the issuer of the different components of the product, or the relevant body, in order to gather the relevant information on these costs.
39. Fair value is the price that would be received to sell an asset or paid to transfer a
liability in an orderly transaction in the principal (or most advantageous) market at the measurement date under current market conditions (i.e. an exit price) regardless of whether that price is directly observable or estimated using another valuation technique.
40. The fair value policy that governs the measurement of the fair value shall set a series of
rules including in the following areas:
(a) governance;
(b) methodology for the calculation of the fair value.
41. Such rules shall aim at outlining a valuation process that:
(a) complies with the applicable accounting standards, in relation to fair value; (b) makes sure that internal pricing models for PRIIPs are consistent with the methodologies, modelling and standards used by the PRIIP manufacturer to value its own portfolio under the hypothesis that the product is available for sale or held for trading; (c) is consistent with the level of complexity of the product and the type of underlying; (d) takes into account the issuer credit risk and the uncertainty about the underlying; (e) sets the parameters to identify an active market in order to avoid risk mispricing that could lead in extreme cases to significantly inaccurate estimates;

67
(f) maximises the use of relevant observable market inputs and minimizes the use of unobservable inputs.
42. The fair value of a structured product shall be determined on the basis of:
(a) market prices, if available or efficiently formed; (b) internal pricing models using as an input market values which are indirectly connected to the product, derived from products with similar characteristics (comparable approach); (c) internal pricing models based on inputs which are not derived directly from market data for which estimations and assumptions must be formulated (markto-model approach).
43. If the fair value cannot be derived from market prices, it shall be calculated using a
valuation technique that is able to represent properly the different factors affecting the product payoff structure making maximum use of market data.
44. A valuation technique shall consider the following according to the complexity of the
product:
(a) the use of recent arm’s length market transactions between knowledgeable, professional counterparties; (b) reference to the current market price of another instrument that is substantially the same; (c) the use of an appropriate discounted cash-flow model where the likelihood of each cash flow is determined using an appropriate model of asset price evolution.
45. In the case of subscription products, the fair value must be calculated on the date when
the product terms are determined. This valuation date shall be close to the beginning of the subscription period and a criterion to update cost information, in case of long offering periods or in case of high volatility in the markets, has to be defined.
46. If preliminary terms are used, costs need to be calculated by using the minimum terms
of the product.
47. If variable subscription prices are used, a procedure on how to incorporate and disclose
the cost effect of the varying subscription price has to be defined.
III. List of Costs of Insurance-based investment products
Costs to be disclosed
One-off costs
48. A one-off cost is an entry and exit cost which includes initial charges, commissions or
any other amount paid directly by the retail investor or deducted from the first payment or from a limited number of payments due to the retail investor or from a payment upon redemption or termination of the product.

68
49. One-off costs are borne by an insurance-based investment product, whether they
represent expenses necessarily incurred in its operation, or the remuneration of any party connected with it or providing services to it.
50. One-off costs include, but are not limited to, the following types of entry costs and
charges that shall be taken into account in the amount to be disclosed for insurancebased investment products:
(a) structuring or marketing costs;
(b) acquisition, distribution, sales costs;
(c) processing/operating costs (including costs for the management of the insurance cover); (d) cost part of biometric risk premiums referred to in point 60; (e) costs of holding required capital (up front part to be disclosed insofar as they are charged). Recurring costs
51. Recurring costs are payments regularly deducted from all payments from the retail
investor or from the amount invested or amounts that are not allocated to the policyholder according to a profit sharing mechanism.
52. The recurring costs include all types of costs borne by an insurance-based investment
product whether they represent expenses necessarily incurred in its operation, or the remuneration of any party connected with it or providing services to it.
53. The following list is indicative but not exhaustive of the types of recurring charge that
shall be taken into account in the amount to be disclosed:
(a) structuring or marketing costs;
(b) acquisition, distribution, sales costs;
(c) processing / operating costs (including costs for the management of insurance cover); (d) cost part of biometric risk premiums referred to in point 60; (e) other administrative costs; (f) costs of holding capital (recurring part to be disclosed insofar they are charged); (g) Any amount implicitly charged on the amount invested such as the costs incurred for the management of the investments of the insurance company (deposit fees, costs for new investments, etc.); (h) payments to third parties to meet costs necessarily incurred in connection with the acquisition or disposal of any asset owned by the insurance-based investment product (including transaction costs as referred to in points 7 to 23).
54. Where an insurance-based investment product invests a part of its assets in UCITS,
AIFs, in a PRIIP other than UCITS of AIFs or in an investment product other than a PRIIP points 5(l), 5(m) and 5(n) shall be applied respectively. Cost disclosure of the biometric risk premium of insurance based investment products Costs part of biometric risk premiums

69
55. Biometric risk premiums are those premiums paid directly by the retail investor or
deducted from the amounts credited to the mathematical provision or from the participation bonus of the insurance policy, that are intended to cover the statistical risk of benefit payments from insurance coverage.
56. The fair value of biometric risk premiums is the expected present value (according to
the interest rates referred to in point 72(a)) of the future benefit payments from insurance coverage taking into account:
(a) best estimate assumptions on these benefit payments derived from the individual risk profile of the portfolio of the individual manufacturer, and (b) other payoffs related to insurance cover (rebates on biometric risk premiums paid back to the policy holders, increase of benefit payments, reduction of future premiums, etc.) resulting from profit sharing mechanisms (legal and/or contractual).
57. Best estimate assumptions on future benefit payments from insurance coverage shall be
set in a realistic way.
58. The estimated future benefit payments shall not include prudency margins or costs for
the management of the insurance cover.
59. For manufacturers within the scope of Directive 2009/138/EC these best estimate
assumptions shall be consistent with the respective assumptions used for the calculation of the technical provisions in the Solvency II balance sheet.
60. The cost part of biometric risk premiums is the difference between biometric risk
premiums charged to the retail investor referred to in point 55 and the fair value of the biometric risk premiums referred to in points 56;
61. Instead of the cost part of biometric risk premiums referred to in point 50(d) and point
53(d) the PRIIP manufacturer may include the full biometric risk premiums in the oneoff, or recurring costs.
PART 2
Summary cost indicators and Compound effect of the costs
I. Summary cost indicators
Summary cost indicator – summary cost indicator
62. The summary cost indicator of the PRIIP is the reduction of the yield due to total costs
calculated according to points 71 to 73.
63. For the calculation of the summary cost indicator the costs to be disclosed referred to in
point 73 shall be the total costs. This shall equal for investment funds the sum of the costs as referred to in points 1 and 2 plus the sum of the costs as referred to in points 4 and 6; for PRIPs other than investment funds, except PRIIPs referred in Annex IV point 12, the sum of the costs as referred to in points 27 and 28 plus the sum of the costs as referred to in point 32 and 33; for PRIIPs referred in Annex IV point 12, the sum of the costs as referred to in points 35 and 36; and for insurance-based investment

70 products, the sum of the costs as referred to in points 48 and 49 plus the sum of the costs as referred to in points 51 and 52. The total costs shall also include exit penalties, if relevant. One-off costs and entry/ exit cost ratios
64. The entry and exit costs ratio of the PRIIP shall be the reduction of the annual yield
due to entry and exit costs calculated according to points 71 to 73.
65. For the calculation of the entry and exit costs ratio the costs to be disclosed referred to
in point 73 shall for investments funds be the entry and exit costs according to points 1 and 2; points 27 and 28 for PRIPs other than investment funds, except PRIIPs referred in Annex IV point 12; point 35 for PRIIPs referred in Annex IV point 12; and points 48 and 49 for insurance-based investment products. Exit costs shall also include exit penalties, if relevant. Recurring costs, portfolio transaction costs and insurance costs / other recurring costs ratios
66. The portfolio transaction costs, insurance costs and other recurring costs ratio of the
PRIIP shall be the reduction of the annual yield due to portfolio transaction costs and other recurring costs calculated according to points 71 to 73.
67. For the calculation of the portfolio transaction costs ratio and the insurance costs ratio:
a) For the calculation of the portfolio transaction, the costs to be disclosed referred to in point 73 shall be the portfolio transaction costs according to points 7 to 23 for investment funds, point 29(c) for PRIPs other than investment funds, except PRIIPs referred in p Annex IV point 12, and point 53(h) for insurance based investment products. b) For the calculation of the insurance costs ratio, the costs to be disclosed referred to in point 73 shall be the insurance costs according to points 60 and 61 for insurance based investment products.
68. The other recurring costs ratio shall be the reduction of the annual yield due to other
recurring costs that is calculated as the difference between the summary cost indicator according to point 62 and the sum of the one-off costs ratio, according to point 64, plus portfolio transaction costs ratio, according to point 67 (a), plus insurance costs ratio, according to point 67 (b), plus the incidental costs ratios, according to point 69. Incidental costs and incidental costs ratios (performance fees and carried interests ratio)
69. For the calculation of the performance fees ratio, the cost to be disclosed referred to in
point 73 shall be the portfolio incidental costs according to point 6 (a) for investment funds. For the calculation of the carried interests ratio, the cost to be disclosed referred to in point 73 shall be the portfolio incidental costs according to point 6 (b) for investment funds.

71
70. The ‘ongoing costs’, ‘performance fees’ and ‘carried interests’ as referred to in Annex
VII are respectively the ‘recurring costs’, ‘performance fees ratio’ and ‘carried interests ratio’ as referred to in the present Annex VI and in Article 5. Calculation of summary cost indicator
71. The summary cost indicator shall be calculated as the difference between two
percentages i and r where r is the annual internal rate of return in relation to gross payments by the retail investor and estimated benefit payments to the retail investor during the recommended holding period and i is the annual internal rate of return for the respective cost free scenario.
72. The estimation of future benefit payments under point 71 shall be based on the
following assumptions:
(a) Except for PRIIPs as referred to in point 12 of Annex IV, the annual internal rate of return, i.e. the performance, of the PRIIP shall be calculated applying the methodology and the underlying hypothesis used for the estimation of the moderate scenario from the performance scenarios section of the key information document; (b) The benefit payments shall be estimated under the assumption that all costs included in the total costs according to point 63 are deducted; (c) For PRIIPs as referred to in point 12 of Annex IV, the performance shall be 3%
73. Calculations for the cost free scenario mentioned in point 71:
(a) For the calculation of i either gross payments by the retail investor from the calculation of r shall be reduced by the costs to be disclosed or the projected benefit payments to the retail investor from the calculation of r shall be increased under the assumption that the amounts of the costs to be disclosed had additionally been invested. Then i is the annual internal rate of return in relation to these adjusted payments by and to the retail investor. (b) Where costs to be disclosed can be expressed as a constant percentage of the value of the assets they may be disregarded in the calculation described in point 73(a) and instead be added to i afterwards. Specific requirements for PRIPs other than investment funds:
74. For the purpose of the calculation of the cost free scenario mentioned in point 73 for
PRIPs other than investment funds, gross payments by the retail investor from the calculation of r, as referred to in point 73, shall be reduced by the costs to be disclosed. Specific requirements for insurance-based investment products:
75. For the purpose of the calculations described in points 71 to 73 for insurance-based
investment products it shall be assumed that no payments resulting from insurance coverage occur during the holding period i.e. the calculation of the summary cost indicator shall be solely based on estimated endowment benefit payments.
76. To the extent recurring and one-off costs are covered by explicit costs that are a
fixed part of the premium calculation of the product the calculation of recurring and one-off costs shall be based on these explicit costs.

72
77.Profit participation for insurance based investment products:
(a) When calculating recurring and one-off costs for insurance-based investment products amounts retained from the investment return through profit sharing mechanisms shall be considered as costs; (b) Where a part of costs is returned to policy holders by separate cost bonuses this shall be considered as a cost rebate that reduces cost deductions provided:
(i) The cost bonuses are declared separately from other parts of the participation bonus and are intended for refunding parts of the costs by the contractual terms of the product. (ii) The insurer can substantiate on the basis of sound actuarial methods that expected future cost bonuses are covered by expected future profits that result from prudent assumptions on future costs. Calculation of ratios Anti-double counting principle
78. If one type of cost is covered by two or more types of costs as referred to in the
present Annex, this type of cost shall only be accounted for once in the calculation of the indicators (ratios) which are based on it. Other specifications 79.These ratios shall be expressed as a percentage to two decimal places.
80. These ratios shall be calculated at least once a year.
81. These ratios shall be based on the most recent cost calculations which the PRIIP
manufacturer has determined. Without prejudice to point 78, the costs are assessed on an ‘all taxes included’ basis. As for investment funds:
(a) a separate calculation shall be performed for each share class, but if the units of two or more classes rank pari passu, a single calculation may be performed for them; (b) in the case of a fund which is an umbrella, each constituent compartment or sub-fund shall be treated separately for the purpose of this Annex, but any charges attributable to the fund as a whole shall be apportioned among all of the sub-funds on a basis that is fair to all investors.
82. Apart from the first calculation for a new PRIIP, and if not stated otherwise, these
ratios shall be calculated at least once a year, on an ex-post basis. Where it is considered unsuitable to use the ex-post figure because of a material change, an estimate may be used instead until reliable ex-post figures reflecting the impact of the material change become available.
83. The ex-post figures shall be based on recent cost calculations which the PRIIP
manufacturer has determined on reasonable grounds to be appropriate for that purpose. The figures may be based on the costs set out in the PRIIP’s statement of operations published in its latest annual or half-yearly report, if this is sufficiently recent; if it is

73 not, a comparable calculation based on the costs charged during a more recent 12- month period shall be used instead.
84. Information about the ratios that were applicable during previous years / periods
shall be published at the location which is specified in the key information document as the general source of further information for investors who require it.
85. Where the costs attributable to an underlying UCITS or AIF are to be taken into
account:
(a) the cost indicator of each underlying UCITS or AIF is pro-rated according to the proportion of the PRIIP’s net asset value which that UCITS or AIF represents at the relevant date (being the date at which the PRIIP’s figures are taken); (b) all the pro-rated figures are added to the total cost figure of the investing PRIIP itself, thus presenting a single total. Calculation methodology for new PRIIPs
86. In place of ex-post data, estimates shall be used in the calculation of the different
types of costs. Such estimates shall be carried out by adopting as proxies either a comparable PRIIP or a peer group.
87. For PRIIPs which charge a fixed all-inclusive fee, that figure shall be used
(provided it includes all costs to be presented under the PRIIPs cost disclosure requirements).
88. For PRIIPs which set a cap or maximum on the amount that can be charged, (and
provided it includes all costs to be presented under the PRIIPs cost disclosure requirements) where instead that figure shall be used so long as the PRIIP manufacturer gives a commitment to respect the published figure and to absorb any costs that would otherwise cause it to be exceeded.
89. If, in the PRIIP manufacturer’s opinion, expressing a figure to two decimal places
would be likely to suggest a spurious degree of accuracy to investors, it shall be sufficient to express that figure to one decimal place.
90. The PRIIP manufacturer shall ensure that the accuracy of the estimated figure is
kept under review. The PRIIP manufacturer shall determine when it is appropriate to begin using ex-post figures rather than an estimate; but in any case it shall, no later than 12 months after the date on which the PRIIP was first offered for sale in any Member State, review the accuracy of the estimate by calculating a figure on an expost basis.
II. Compound effect of the costs
Common requirements to all types of PRIIPs
91. The table(s) referred to in Article 5 of this Regulation shall contain an indication
of the total costs in monetary and percentage terms for the case that the retail investor invests, respectively 10 000 € (for all PRIIPs except regular premium insurance-based investment products), or 1 000 € yearly (for regular premium insurance-based investment products) during different holding periods, including the recommended

74 holding period. The holding periods to be shown are those referred to in points 14 to 16 of Annex IV. Where a product is considered not to have an alternative liquidity facility promoted by the PRIIP manufacturer or a third party, or where there is an absence of liquidity arrangements, or for those PRIIPs as referred to in Annex IV point 12, the aforementioned indication of costs may be shown only at maturity or at the recommended holding period.
92. Where the currency of the PRIIP is not in Euros, an amount of a similar magnitude
to those set out in point 91 and which is cleanly divisible by 1000 shall be used.
93. The total costs shall include one-off, recurring and incidental costs, and, if
relevant, exit penalties.
94. Exit penalties are to be distinguished from other exit costs which have to be paid in
any case and therefore always need to be included in the one-off costs.
95. The relevance of exit penalties depends on the holding period of the investment
and the exact moment when the product is cashed in. Exit penalties are not relevant if the investment is kept for the recommended holding period.

75
ANNEX VII
PRESENTATION OF COSTS
The Reduction in Yield (RIY) shows what impact the total costs you pay will have on the investment return you might get. The total costs take into account one-off, ongoing and incidental costs. The amounts shown here are the cumulative costs of the product itself, for three different holding periods. They include potential early exit penalties. The figures assume you invest [€10 000 (OR €1000 each year for regular premium PRIIPs)]. The figures are estimates and may change in the future.
Table 1 Cost over time
The person selling you or advising you about this product may charge you other costs. If so, this person will provide you with information about these costs, and show you the impact that all costs will have on your investment over time.

76
Table 2 Composition of costs
The table below shows:
 The impact each year of the different types of costs on the investment return you might get at the end of the recommended holding period  What the different cost categories mean. On the same basis, the additional impact of your insurance premium payments (equivalent to the estimated value of insurance benefits) is [x%]. Details of the insurance benefits such as death benefits are in the section “what is this product”.

78
4. Accompanying Documents
4.1 Impact Assessments
4.1.1 Impact Assessment for Article 8 (5)

Procedural issues and consultation of interested parties
According to ESAs’ Regulation, the ESAs conduct analysis of costs and benefits when drafting regulatory technical standards. The analysis of costs and benefits is undertaken according to an Impact Assessment methodology. The draft RTS and its impact assessment are subject to public consultation. In preparing these draft RTS, the ESAs have conducted two separate public consultations by means of Discussion Papers.
Description of market
The PRIIPs market is broad and diverse, both at the EU level and between different national markets. Globally, the larger part of the market in terms of assets under management is investment funds, mostly in the form of UCITS; insurance-based investment products (either unit-linked, or ‘traditional’ or ‘with profits’ in type); and structured products. Estimates from 2009 put the UCITS market at around 9 trillion EUR, and insurers assets at around 6 trillion EUR, and retail structured products at around 0.5 trillion EUR (it is unclear whether these figures included structured deposits). 10 There is some difficulty assessing the end proportions of retail assets under management under these different vehicles: UCITS will often be underlying investments for insurance-based investment products, irrespective of whether these latter are directly unit-linked or traditional in form, so there will be some double counting, while the insurance figures are not broken down according to assets under management earmarked for retail investors. In addition EFAMA estimates show perhaps only 33% of UCITS sales can be directly classified as retail. Overall the market might be considered to be around 10 trillion EUR in size. While there can be expected to be some evolution of this picture over time, for instance an increased allocation of assets to innovative instruments, such as structured products, in a ‘low yield’ environment, the broad rankings of vehicles according to assets under management are

10 http://ec.europa.eu/internal_market/finservices-retail/docs/investment_products/20120703-impactassessment_en.pdf, p.

79 unlikely to have changed significantly since 2009 given the relatively large scale of the UCITS and insurance sectors compared to others. Having said this, national markets have developed in different ways, such that the balance between investment vehicles can vary according to local markets: in some, retail structured products may be more prevalent, or direct sales of UCITS, or unit-linked contracts as opposed to traditional insurance investment vehicles. This heterogeneity also reflects the balance of features offered in vehicles: for instance, guarantees, whether third party or not, are more commonly offered in some markets than others, irrespective of the investment vehicle. PRIIPs are also heterogeneous in regards their features, though, as PRIIPs, they of course share many features too, and by definition all target accumulation and investment aims. This heterogeneity reflects such factors as the different time horizons they are designed for (from some instruments that can be suitable for very short period investments, for weeks or months, to products designed for accumulating savings over decades), the different degrees of diversification and risk spreading they seek to offer, and differences in terms of the nature and extent of capital protection provided. For PRIIPs that are insurance based, biometric risk cover is also included. The heterogeneity is important when considering the impacts of different options for standardised KIDs.
3. Problem definition
According to Article 8(5), the ESAs shall develop draft RTS specifying:
A. Details of presentation and content of information contained in a key information document (KID) (following the sections set out in Article 8 (3)); B. the methodology to be used for the summary risk indicator and performance scenarios; and
C. the methodology to be used for the summary cost indicators.
The scope of these draft RTS is therefore broad, but not without significant limits: specific rules already apply in the Regulation on the length of the document, the use of plain language, the headings to use in the document and the sequence that these need to follow, and the substantial information elements to be included in the different sections. The general purpose of the draft RTS is to standardise the KID to the greatest degree, to aid consumers in comprehending and comparing different PRIIPs. This general purpose is determined by the level one Regulation and so to a great extent impacts that are related to such standardisation in general are not relevant for the assessment of different options for such standardisation. By contrast, the assessment of different options in regards the draft RTS should focus on the differential impacts of these options.

80
The level one Regulation was accompanied already by an impact assessment, focused on the specific problems and possible solutions arising in relation to product information (‘disclosure’). 11 The key problem drivers identified were:
 the emergence of retail investment products of similar economic nature but different legal forms;  a patchwork of regulation applying to retail investment products; and  unmitigated asymmetries of information. These drivers led to disclosures that were insufficient or difficult to comprehend and compare, and an uneven level playing field between product manufacturers. The impact assessment set out evidence that these in turn caused or contributed to individual consumer detriment through missales, a decline in confidence in the investment markets more generally, and thereby reduced capital-market efficiency. The impact assessment recognised however that there could be multiple determinants of mis-sales, not solely poor product disclosure. These include inadequate or misleading advice, and issues related to unmitigated conflicts of interest. The specific problems drivers and problems that are to be considered in respect of the level two draft RTS are in the context of this broader picture, as explored in the level one impact assessment. The ESAs have assessed the areas where there is significant discretion at level two as the areas of risk disclosure (notably, the summary risk indicator), the treatment of rewards (performance scenarios), and the information on costs and summary cost indicators. The ESAs have identified the problems that specifically need to be addressed in these areas as follows:
Problem driver Problem Detail
Complexity/diversity of risks
Reduced capacity to understand and compare risks  Risks opaque  Synthesising risks difficult  Risks difficult to compare Uncertainty of rewards Reduced capacity to understand and compare rewards  Rewards unclear  Lack of balanced perspective on rewards, including potential downside as well as upside  Possible rewards difficult to compare between PRIIPs Complexity / diversity of costs Reduced capacity to understand and compare costs  Overall impact/scale of costs difficult to understand  Unclear impact of costs over time  Difficult to compare costs

11 http://ec.europa.eu/internal_market/finservices-retail/docs/investment_products/20120703-impactassessment_en.pdf.

81 for PRIIPs, especially where these have different cost structures Baseline scenario When analysing the impact from proposed policies, the impact assessment methodology foresees that a baseline scenario is applied as the basis for comparing policy options. This helps to identify the incremental impact of each policy option considered. The aim of the baseline scenario is to explain how the current situation would evolve without additional regulatory intervention. The baseline scenario is based on the situation where the PRIIPs Regulation applies, but where there is no RTS to further specify Article 8(3). Individual PRIIP manufacturers would each be liable to develop risk, reward and cost disclosures. Some coordination amongst manufacturers, for instance through trade bodies, to develop possible ‘soft law’ approaches might be expected, but cross-sectoral and to a degree cross-market consistency, would not be likely. The specific liability for each PRIIP manufacturer to develop a risk indicator, performance scenarios and cost indicators could be expected to drive considerable costs in the aggregate across markets, sectors and manufacturers. The costs could be disproportionate in application on different PRIIP manufacturers. Trade body (‘soft law’) activities would not fully mitigate these costs. The lack of consistency in content – even where there is to a degree strong consistency in format – of the KID could lead to detriment amongst retail investors where they rely on the document to make comparisons between PRIIPs unaware of or not taking sufficient notice of differences in the methodologies used for the risk, performance and cost indicators. This could in turn lead to reduced reliance and trust in the KID amongst retail investors and distributors, significantly undermining its ability to achieve benefits for retail investors.
4. Objectives
As noted above, in order to ensure consistent application of Article 8 (3), the ESAs are empowered under Article 8 (5) to specify the details of the presentation and the content of each of the elements of information in the KID, on the methods for the risk and reward information, and the methods for the costs information. When developing the draft regulatory technical standards, the PRIIPs Regulations sets out that the ESAs shall take into account the various types of PRIIPs, the differences between them and the capabilities of retail investors as well as the features of the PRIIPs so as to allow the retail investor to select between different underlying investments or other options provided for by the product, including where this selection can be undertaken at different points in time, or changed in the future.

82
In ensuring consistent application of Article 8 (3), the ESAs identified the following general objectives, taking into account the problems set out above in section 3, whereby consistent means for achieving improvements to comparability and comprehensibility can be found:
Problem driver Objective Detail
Complexity/diversity of risks
Improve comparability and comprehensibility of risks  Comparable risk indicator  Indicator makes overall impact of risks clear  Risk indicator and accompanying narrative Uncertainty of rewards Improve comparability and comprehensibility of rewards  Balanced perception of upside and downside  Understanding of nature/shape of reward profile  Comparability between very different payoff structures Complexity/diversity of costs Improve comparability and comprehensibility of costs  Understanding of combined impact of costs  Clarification of the costs evolution through time  Comparability of costs between very different cost structures  Awareness of how cost structure could apply for particular circumstances of retail investor In assessing options for tackling these broad objectives, the ESAs have identified a number of criteria that are important. These relate both to questions of ‘presentation,’ and to questions, where figures are involved, of methodologies to be used. These criteria also link to the overall objective (consistency in measures to improve comparability and comprehensibility) These criteria were already set out in the first Discussion Paper (JC/DP/2014/02):
Criteria for assessing options for presentation Engaging The presentation of the elements in the KID should be as simple as possible, to be easily understood by retail investors and to engage them more, motivate them to use the KID and increase attention for the KID Understandable This criterion refers to the level of complexity of the information as whether consumers can interpret the information. The information in the KID should be understood by the retail investors with the assumption that the consumer may not have an adviser, distributor or seller on hand to explain the information. Comparable One of the main purposes of the KID is to enable investors to compare investment opportunities with each other. This can be done by an investor by comparing several KIDs but it is also possible to include reference points within the KID that make the information more comparable, for example through the use of benchmarks. Compatible This criterion refers to the compatibility of the proposals for presentation of information with the requirements formulated in the PRIIPs Regulation. All options will need to be

83 strictly compliant with the PRIIPs Regulation.
Balanced presentation
This criterion holds that we should look for a balanced presentation of the different aspects within sections of the KID but also a balanced presentation of the sections of the KID. By this criterion we try to make sure that the upsides and downsides of certain products are balanced in an objective fashion. Coverage of types of PRIIPs The goal is to develop presentational forms that are suitable and applicable to all different types of products that fall into the scope of the PRIIPs Regulation. Given the heterogeneity of PRIIPs in scope, this will be an important challenge. Criteria for assessing underlying methodologies Reliable The information within the KID is reliable where it provides a fair estimate of the actual risks and costs involved. Robust The measurement of risk, reward and costs should not be easy to manipulate. It should be an objective representation of the risk, reward and costs that are being measured. Stable The output of the measurements needs to be relatively stable. It is important that risk or cost indicators are reliable forecasts, and that they are not overly sensitive to relatively minor changes in conditions. Applicable The measurements should be applicable to all types of PRIIPs. Where a measurement (e.g. of historic volatility) is available for some PRIIPs but not available for those without a track record, or might be a misleading measure for some, effective methodologies for combining or synthesizing different measures in an objective way may be necessary. Comparable The measurements should lead to values that are comparable amongst different types of PRIIPs. Discriminatory If it is not possible to differentiate between PRIIPs, a measure loses its purpose. It needs to be clear that a certain measured output is below or above another measured output. Therefore it is important that the indicator provides discriminatory output. Feasible/ Proportional An indicator or measure that is overly sophisticated in relation to the granularity and accuracy of the information included in the KID could be seen as disproportionate. This does not imply that the simplest or least costly option should always be selected. Supervision Will it be possible for regulators to assess whether product manufacturers are complying with any proposed prescribed methodology? These criteria have been used – as appropriate and where relevant – throughout the internal work of the ESAs when identifying and developing options. This impact assessment will concentrate however only on the main aspects of the most material of the policy issues, as identified by the ESAs.
5. Policy issues and options
In respect of each of the broad objectives above, the ESAs have identified the following policy issues where different policy options had a strong potential for materially different costs and benefits. Each policy issue is assessed in depth below. Issue Description Risk 1 Selection of types of risk to include in the SRI/risk and reward section Risk 2 Aggregation of main types of risks Risk 3 The different aspects of the Methodology to get to a Summary Risk Indicator Risk 4 Disclosure of liquidity risk and liquidity profile (liquidity issues) Risk 5 Selection of SRI time frame Risk 6 SRI presentation format Reward 1 Choice of Performance Scenarios Reward 2 Performance Scenarios presentation format

84
Costs 1 Investment Funds / Transaction costs
Costs 2 Investment funds / Performance fees
Costs 3 PRIPs other than investment funds / Disclosure of certain types of costs for structured products (and derivatives and CFDs) Costs 4 Insurance-based investment products / Biometric risk premiums (BRP) and summary cost indicator Costs 5 Insurance-based investment products / Summary indicator for insurance-based investment products Costs 6 General / Summary cost indicator The ESAs have also identified the importance of specifically assessing the impact of proposals related to the above in regards of options for ‘multi-option products’. This is addressed separately below. Policy issue 1: Selection of types of risk to include in the SRI/risk and reward section Policy option 1.1 – No reference to specific types of risk Policy option 1.2 – All types of risk identified in the DP Policy option 1.3 – Circumscribe risks to market, credit and liquidity The PRIIPs Regulation indicates that the KID needs to include a summary risk indicator, supplemented by a narrative explaining its main limitations and the risks which are materially relevant to the PRIIP and which are not adequately captured by the summary indicator. The Regulation does not specify which risks shall be included in the summary risk indicator and which others may be better explained in a narrative. To enable an RTS on a SRI a specification of the most relevant types of risk was needed. For this reason the most relevant risks have been enumerated in the Discussion Paper. Based on the responses it was clear to ensure the section on risk needs to include and highlight at least information on the market, credit and liquidity risks of a PRIIP.
1.1 No reference to specific types of risk
No. Pros Cons
1
Gives manufacturers more freedom to assess which risks are relevant to each PRIIP. Not compliant with L1, since it is not possible to define a methodology for the SRI. 2 Subjective, difficult or impossible to supervise. No consistent market approach expected, preventing comparability. 3
1.2 All types of risk identified in the DP
No. Pros Cons
1 Full accountancy of risks.
Some of those risks can be subsumed into the main selected risks.

85
No. Pros Cons
2
Some of those risks are potentially negligible – there is not an increased level of discrimination stemming from the added complexity of the approach. 3 Not all risks are relevant to all types of PRIIPs/all types of retail investors (e.g. currency risk, only relevant when the investor´s currency is different from the PRIIP´s currency) or are measurable (e.g. legal risk). 4 When it is stated or suggested that all possible risks are included in the SRI, this might create overreliance on the SRI and therefore a possible moral hazard with regards to investment risks. No more information gathered by consumers and overreliance on the KID, which is summarizing product information in the orientation phase, could be a consequence of that.
1.3 Circumscribe risks to market, credit and liquidity
No. Pros Cons
1
Specification of these risks makes it clear which risks are worth being considered. A comparable approach can be established. No accountancy of other potentially relevant risks – SRI underestimates true risk. In any case, narratives will be used to highlight those other risks (still, they won´t help investors assess the level or impact of the risks excluded). 2 According to responses to the DP these are considered the most relevant risks. 3 Some of other risks described in the DP can be subsumed into the main selected risks. Policy issue 2: Aggregation of main types of risk into the SRI Policy option 2.1 – No aggregation – main types of risk shown separately Policy option 2.2 – Aggregation of main types of risk (market, credit and liquidity) Policy option 2.3 – Aggregation of market and credit risks. Liquidity risk displayed through narratives and/or warnings Following the selection of risks to include in the SRI, these risks needed to be measured (policy issue 2.3) and a way to display them together had to be envisaged. A few options were

86 considered. The first option was to display the three types of risk separately from each other. The second option was to aggregate the three types of risk into the SRI and the third option was to solely aggregate market and credit risk and to display liquidity risk separately, by means of narrative and/or warning.
2.1 No aggregation – main types of risk shown separately
No. Pros Cons
1 Easier to implement. Disclosure of separate indicators without an overall summary, is not compliant with L1. 2 Displays the different types of risk enabling investors to find out the origins of risk. Difficult for retail investors to synthesize the impact of the three types of risk, weighing the risks and the consequences thereof is left to the consumer, where they are not so well equipped to assess the impact. 3 By separating the risks it may become more difficult to compare the different PRIIPs with each other based on their risk assessment.
2.2 Aggregation of main types of risk (market, credit and liquidity)
No. Pros Cons
1 Compliant with L1. Complex to weigh the different types of risk.
2
Easier for retail investors to understand the overall risk and to compare products. From the indicator itself it is not clear the weight or contribution of each type of risk However, narratives can be used for this purpose. 3 Liquidity risk difficult to aggregate in the same scale as market and credit risk due to difference in the nature and subjective weighting of this risk.
2.3 Aggregation of market and credit risks. Liquidity risk displayed through narratives/warnings
No. Pros Cons
1 Compliant with L1.
Added complexity especially with regard to the weighing of the two types of risk in the SRI - requires the use of judgment calls. 2 Liquidity risk difficult to quantify and weigh against other risks so better displayed separately. Difficult to figure out the individual risks´ contribution to the overall risk level displayed in the SRI, though narratives may be used for this purpose.

87
Policy issue 3: The different elements of the methodology underlying the calculation of the Summary Risk Indicator Policy issue 3a: Market risk assessment: purely quantitative or combined with qualitative measures Policy option 3a.1 – Purely quantitative Policy option 3a.2 – Purely qualitative Policy option 3a.3 – Combination of quantitative and qualitative measures The three risk types selected for inclusion in the SRI can be assessed in a quantitative or in a qualitative way, or alternatively still in a way that combines the use of some of those quantitative and qualitative measures. With a quantitative assessment reference is made to a methodology that assesses the risk based on numerical data. A qualitative assessment merely bases its categorization on the characteristics (e.g. leverage or capital protection) of products. 3a.1 Purely quantitative No. Pros Cons 1 Several known and generally accepted measures from which to choose depending on the risk definition (“loss” vs “return uncertainty”) Quantitative risk assessment (irrespective of risk definition) is standard market practice. No single measure adequate for all PRIIPs. 2 Objective measures -measure do not require subjective judgment from regulators to assess and compare the impact of different criteria. Some risk features may be difficult to assess quantitatively. 3 Strong preference according to TDP consultation. The computation burden can be heavy for particularly complex products. Dependent on the perception of risk (loss or uncertainty) it may be redundant to require a quantitative assessment for PRIIPs with a full guarantee 3a.2 Purely qualitative No. Pros Cons 1 Relatively easy to assess whether a PRIIP fulfils the characteristics or not It is not clear whether the qualitative characteristics of a PRIIP are a good proxy for the actual market risk of a PRIIP 2 Qualitative assessment is less burdensome for manufacturers as the characteristics for a PRIIP will usually not change over time so Comparison between products may not be sufficiently sophisticated to distinguish products over different classes

88
No. Pros Cons review and revision of a SRI will be limited 3a.3 Combination of quantitative and qualitative measures No. Pros Cons 1 More refined approach, intended to provide increased reliability and discriminatory power More subjective, given the use of some previously defined judgment calls 2 More cost efficient approach, may avoid the need of complex calculations for certain products. May be disputable which products should be assessed on a quantitative basis and which not. 3 More reliable approach in respect of extremely risky products (e.g. products where the investor can lose more than the invested amount). Policy issue 3b: Market risk quantitative measure: level of prescription and data on which is based. Complexity of the methodology. Policy option 3b.1 – Highly standardized methodology based on historical volatility, with an adaptation for non-linear products based on historical simulation (UCITS solution) Policy option 3b.2 – Low standardized methodology (high level principles), based on forward looking models and a downside measure for all products Policy option 3b.3 – Hybrid approach: highly standardized methodology using historical data. The methodology is adapted to the type of product, considering in any case more information about the shape of the distribution than the volatility and using a downside measure Given the different products in scope it is not likely that there will be one methodology to determine the market risk which suits all products in scope. As already included in the UCITS SRRI methodology where for structured UCITS a different approach is applied. Dependent on the methodology selected the method can be more or less standardized. The policy options described in this section are the following:
3b.1 Highly standardized methodology based on historical volatility, with an adaptation for non-linear products based on historical simulation (UCITS solution) No. Pros Cons 1 Volatility is a well understood and well established concept in finance Volatility as a risk measure is not adequate for all PRIIPs (e.g. insurance products and fixed term products in general) 2 Familiar methodology, in use for UCITS since 2011 – easy to implement and easy to supervise History does not repeat itself, i.e., past volatility is not a good estimator of future volatility

89
No. Pros Cons
3 Significant support among TDP respondents
3b.2 Low standardized methodology (high level principles), based on forward looking models and a downside measure for all products No. Pros Cons 1 The VaR approach (or any downside measure) is in line with consumers’ main risk perception (i.e., “loss”) With models and parameters left at the discretion of manufacturers, it is difficult to ensure a common approach and proper comparability 2 Very reliable (all risk factors considered) and applicable to all PRIIPs in scope Some feasibility problems expected namely in regards to products that invest in a big number of underlying investments that can be actively managed. 3 Good support among TDP respondents More difficult to supervise 3b.3 Hybrid approach: highly standardized methodology using historical data. The methodology is adapted to the type of product, considering in any case more information about the shape of the distribution than the volatility and using a downside measure No. Pros Cons 1 High level of standardization ensures robustness of the methodology and facilitates the creation of a risk scale able to discriminate and compare different types of products. High level of standardization reduces flexibility to adapt to certain products or market situations. 2 Allows a less complex analysis for products that have a linear pay-off. History does not repeat itself, i.e., historical data may not be a good base for estimating future behaviour 3 Policy issue 3c: Credit risk assessment Policy option 3c.1 – Based on external credit ratings Policy option 3c.2 – Based on quantitative standardized measures Policy option 3c.3 – Based on internal assessment of the manufacturer Credit risk has been identified as one of the risk measures that need to be included in the SRI. The point with credit risk is that in ordinary circumstances the risk of a credit event is relatively low but the impact of such event could be tremendous on the performance or pay-out of a PRIIP. There are several ways in which credit risk can be measured. The most flexible approach is to allow manufacturers themselves to assess the credit risk that they put on the retail investor

90
(3c.3). Another approach would be to apply quantitative measures based on which credit risk can be assessed (3b.2). Difficulty may be to identify sufficiently measures to proxy the credit risk that a PRIIP has for the retail consumer. The final method is to base the credit risk assessment on the CRA rating that has been given to the PRIIP itself and if such is not the case the CRA rating of the obligor. The Credit Risk Measure takes into account the default probability of the PRIIP/obligor. 3c.1 Based on external credit ratings No. Pros Cons 1 Generally accepted market practice. May be considered as not preventing overreliance on credit ratings, in contrast with recent regulatory trends. 2 Preferred option according to TDP consultation. Not all entities have ratings, an alternative solution needs to be found for those entities facing a comparative disadvantage. 3 Robust approach that enables comparisons across different manufacturers. 4 Avoids reliance on internal assessment. 5 unequivocal/workable proposal to supervise. 3c.2 Based on quantitative standardized measures No. Pros Cons 1 Potentially easier to concatenate with (quantitative) market risk measure. Credit spreads and CDS spreads can be volatile, unstable, and reflecting market developments not necessarily credit related. Solvency capital ratios should be considered as an option when the regulation is reviewed. 2 Potentially more reactive to changing market circumstances. Not available for all entities. 3 No objective criteria or methodology (not based on market data) currently in use in the markets (work in progress in other regulatory bodies). 3c.3 Based on internal assessment of the manufacturer No. Pros Cons 1 Potentially more reliable as it can lean on more features and take into account current market situations No consistent approach expected, not allowing comparability.

91
No. Pros Cons
2
Potentially quicker to react to changing market circumstances Increased onus on supervision 3 May pose some conflicts of interests‘ problems (and possibly credit risk underestimation problems), considering that the manufacturer is often the obligor whose solvency needs to be assessed. Policy issue 4: Disclosure of liquidity risk and liquidity profile (liquidity issues) Policy option 4.1 – No reference to liquidity issues in the risk (and reward) section. All liquidity related information displayed under the section “How long should I hold it and can I take money out early? (”How long…?”) Policy option 4.2 – Information in the risk (and reward) section limited to liquidity risk. Liquidity profile separately explained under the section “How long …?” Policy option 4.3 – Warning in the risk (and reward) section warning about potential lack of accessibility to the amount invested before maturity or significant limits. Full description of liquidity profile (disinvestment procedures and costs) under the section “How long …?”. Liquidity risk has been defined as one of the main risks to be included in the KID. It is important to distinguish liquidity risk from liquidity profile (laying out the contractually established limits on accessibility to the amount invested in the product before maturity and early redemption costs), which the retail investor is acquainted with from the moment he buys the product. The liquidity risk refers to unforeseen deviations to this liquidity profile, i.e., potential difficulties to have access to the amount invested in the conditions defined in the contract. This section deals with the policy issues related to this specific liquidity issues which is the combination of liquidity risk and liquidity profile.
4.1 No reference to liquidity issues in the risk section, only under “How long ….?” Section
No. Pros Cons
1
Easy to implement. All information about disinvestment details in the same section Liquidity has been considered a major risk type, worth for as inclusion in the risk
section
2 Treat together issues that are interrelated
Lack of accessibility to the money (without significant costs), should that need arise, is a potential source of loss for the investor that would not be referred to in the risk section. 3 Limited accessibility to the money has an impact on the market risk the investor may face - he may not be able to react to adverse market changes, or changes in personal circumstances

92
4.2 Information in the risk section limited to liquidity risk. Liquidity profile under “How long…?”
section.
No. Pros Cons
1
Although the risk profile may limit accessibility to money, it is known in advance (a feature of the product) and hence, it may be considered not a risk, in the sense that there is no uncertainty regarding it. Lack of accessibility to the money (without significant costs), should that need arise, is a potential source of loss for the investor that would not be referred to in the risk section. 2 Allows full description of the liquidity profile and disinvestment procedures. Limited accessibility to the money has an impact on the market risk the investor may face - he may not be able to react to adverse market changes, or changes in personal circumstances 3
4.3 Warning in the risk section for accessibility problems. Full description of liquidity profile
under “How long…?” section
No. Pros Cons
1
Advises investors about possible potential problems in accessing their accessibility to money, should they need it, in a standardized format that helps passing that message along. Liquidity profile may not technically qualify as a risk, if risk presumes uncertainty. 2 Treat together issues that are interrelated and not well discriminated amongst by retail investors. 3 Allows for detailed and flexible explanation of liquidity profile in the corresponding
section
Policy issue 5: Selection of SRI time frame
Policy option 5.1 – Short and standardized holding period for all products Policy option 5.2 – Multiple time frames Policy option 5.3 – End of maturity/Recommended holding period Calculating the risk for a certain product is highly dependable of what timeframe is chosen over which the risk will be calculated. The selection of a time frame is even more important for a quantitative assessment of market risk. Not only is this relevant for the period of data to be collected, but also because the risks might be significantly different for the same product at different time frames.

93
Allowing for flexibility in timeframes uses may result in difficulties comparing different risk assessment with each other. Since the assessment was that a quantitative market risk assessment should be the basis of the SRI, a timeframe needs to be selected in order to calculate the SRI. Different options have been considered on what would be a suitable timeframe to use for the scope of products under PRIIPs.
5.1 Short and standardized holding period for all products
No. Pros Cons
1
Easier to implement, notably in respect of
(simpler) drift assumptions
Short term risk is a biased estimator of longer term risk. For a lot of products/investors with a buy and hold strategy risk matters at maturity only 2 Easier comparison for different products (standard holding period for this purpose) 3 Adapted to those investors that are not long term investors or may be oriented to restructure their portfolios according to market evolution.
5.2 Multiple time frames
No. Pros Cons
1
Full picture of risk throughout the PRIIP´s life Harder to implement, maintain and supervise 2 Comparability ensured at more than one point in time Harder to understand for retail investors 3
5.3 End of maturity/Recommended holding period
No. Pros Cons
1
For a lot of products/investors’ types investors only risk at maturity matters as they have a buy and hold strategy. Easy concept to understand. Requires an additional step in the methodology to make the different time frames comparable to each other. 2 Preferred option according to TDP consultation Varying risk at different points in time not accounted for. Although this limitation of the indicator will be safeguarded through a narrative, the investor is not fully aware of the losses (which are not estimated) he may face before maturity (or the recommended holding period), should he want to disinvest.

94
Policy issue 6: SRI presentation format
Presentation of the SRI instead of other formats tested in CT I and CT II Policy option 6.1 – Simple graphical presentation of the scale Policy option 6.2 – Add other informative elements (reference to reward, narratives for relevant risks) Policy option 6.3 – Add images or other visual elements With regards to the SRI presentation in the first Consumer Testing phase, four very different variants were tested. In the second Consumer Testing phase three different variants of the variant that were best tested in phase I were tested. It appears from consumer testing that objective understanding and comparison, but also subjective preferences of consumers are best met by selecting a simple scale presentation of risk, aggregating different risk factors.
6.1 Simple graphical presentation of the scale
No. Pros Cons
1
Straightforward, intuitive, simple
(numerical) message
No contribution breakdown of different types of risk, (though narratives can be used for that purpose) or other complementary pieces of information 2 Strongly endorsed approach according to CT results Potentially less appealing than images for certain investors 3 Similar to SRRI presentation with which investors are acquainted
6.2 Add other informative elements (reference to reward, narratives for relevant risks,…)
No. Pros Cons
1
Baseline numerical message complemented with other useful information Potentially more confusing message leaving investors unaware as to how to integrate the information about risks displayed in narratives 2 Products more finely differentiated from each other The allusion to reward, in the context of the risk indicator, is not easily seized by all investors (cf. CT result) 3
6.3 Add images or other visual elements
No. Pros Cons
1
Without prejudice to the main numerical message, visual elements are good at capturing investors’ attention to particular product features Some images create inappropriate analogies with other symbols or scales

95
Policy issue 7: Choice of performance scenarios Policy option 7.1: Full manufacturers´ discretion Policy option 7.2: Highly prescribed scenarios based on probabilities or pre-established assumptions Policy option 7.3: Hybrid approach based on principles, including the reasonability of the scenarios (manufacturers´ choice within the limits or guidelines in the RTS) Policy option 7.4: Prescriptive methodology based on the Distribution of Returns required by the calculations for the risk indicator. There are different approaches one could take to prescribe how the performance scenarios shall be calculated. The most flexible approach is to allow manufacturers full discretion to display three (or more performance scenarios in the KID (policy option 7.1). Clearly this would lead to performance scenarios that are not comparable. Furthermore, there is the risk that too optimistic performance scenarios are being displayed which may mislead investors. Another approach is to prescribe the specific assumptions for the risk premium, based on which the performance scenarios should be calculated (policy option 7.2). This would lead to comparable assumptions for all PRIIPs in scope. But is, given the broad range of PRIIPs and the even broader range of underlying investments, likely to result in a lot of different calculation methods for different product types. Furthermore the specific product aspects should be reflected in the calculation of the returns, therefore a standardized methodology is probably providing outcomes which are far removed from outcomes in reality calculated with a better fit model. The third approach is to have an approach which is more based on principles (hybrid). This approach sets the boundaries for the performance scenarios but allows for the different types of PRIIPs to display their different product characteristics in the performance scenarios. The final approach is to use the Distribution of Returns (DoR) of the PRIIP generated for the MRM, which is prescribed and to also prescribe what percentiles should be selected in order to represent the performance scenarios. One could argue that the distribution of returns are already required by the calculation method chosen for the SRI. Here no assumptions would need to be made and therefor prescribed on the risk premiums for different assets (as in policy option 7.2). Preventing to prescribe assumptions to be made on market developments and no straightforward or non-contentious methodology is available to do so. Policy option 7.1: Full manufacturers’ discretion No. Pros Cons 1 Flexibility, ability to focus on specific product features and market situations Lack of robustness. Risk of abuse by manufacturers (showing too favourable potential results) 2 Only face-comparability, a presentation can be prescribed, but the underlying assumptions and methodologies used are not comparable.

96
No. Pros Cons
3
Not compliant with L1 which1 that requires a methodology to be set.
Policy option 7.2: Highly prescribed scenarios based on probabilities or pre-established assumptions No. Pros Cons 1 Robust: it prevents manufacturers from selecting too favourable or biased scenarios Methodologies to forecast the future are difficult to prescribe and are necessarily based on product specific assumptions. 2 Comparable results for different products and manufacturers Pre-established scenarios, not based on estimations, may not be applicable or relevant for all products. May even result in misleading performance scenarios. 3 In the case of a probabilistic approach, there is added value for investors from the information about likelihood of different scenarios (reduce asymmetry of information) Easier to supervise compliance (methodology is more clear), though expertise on modelling may be required Policy option 7.3: Hybrid approach based on principles, including the reasonability of the scenarios (manufacturers’ choice, within the limits or guidelines in the RTS) No. Pros Cons 1 The principles or guidelines increase robustness, preventing manufacturers from selecting too favourable or biased scenarios, particularly those requiring reasonability assumptions Level of prescription may not be sufficient to enable manufacturers to have clear guidelines of what is expected 2 It may be proportional for certain products allowing different approaches to select scenarios for different products May lead to a more meaningful but however less comparable results compared to a prescribed methodology. 3 Should the experience advise further standardization, supervisors may issue additional guidelines to detail a more concrete methodology, at least for certain products. Policy option 7.4: Prescriptive methodology based on the Distribution of Returns required by the calculations for the risk indicator. No. Pros Cons 1 The comparability of the performance scenarios is ensured by prescribing according to the current methodology, since Retail investors will not learn from the performance scenarios what type of distribution the possible future returns

97
No. Pros Cons all PRIIPs for which historical data or a benchmark is available will use the same methodology. The basis of the methodology is comparable to the MRM-calculation have. By picking a percentile no information is given on the skewness and kurtosis of the distribution of returns, meaning no information is or can be given on how likely the outcome will occur (i.e. probabilities). 2 It is a straightforward and relatively simple methodology, which provides certainty to market parties in how to fulfill the requirements of the regulation. The tail of the distribution is to some extent neglected by choosing the 10th and 90th percentile. So the outcome could be even worse or better than displayed in the performance scenarios. Since outcomes before the 10th and after the 90th percentile are not represented, however the maximum loss is required to be mentioned in the narratives. 3 The performance scenarios show the risk and reward trade-off to the retail investor and provide possible future outcomes of the product (similar to the MRM). The methodology holds that, due to the reliance on historical data, that (especially for the Category II PRIIPs) the scenarios are heavily impacted by their past returns. This is less significant for category III products since these products are simulated via Bootstrapping. This holds that for Category II products where a perfectly normal distribution could be established and the RHP is equal to the period of available historical data that is used, the moderate scenario will be representing the average past performance. 4 Products that are automatically assigned to a risk class would have the burden of calculating a distribution of returns as in addition. 5 Products where no sufficient data is available will have to provide a best estimate, for there is no data available to calculate from. This could be based on policy option 1 to further mitigate the level of discretion of the manufacturer. Policy issue 8: Intermediate periods Policy option 8.1: Selection of intermediate periods from the distribution of returns. Policy option 8.2: Extract from the selected scenarios at RHP the underlying paths Policy option 8.4: Best estimate for the intermediate periods in the KID

98
There are different approaches which could be considered in regards to disclosing intermediate periods. This largely depends on which methodology for the performance scenarios is chosen. For products that are linear products with –close to- normal distributions the intermediate periods could be selected from the Distribution of returns. This is because we are assuming for the market risk measure that all market assets follow a ‘time homogenous process’. This holds that the distribution of returns is the same regardless of when I start or stop measuring. However for nonlinear products more simulations and calculations are needed to find relevant values to present as possible outcomes of the intermediate periods (before RHP). Policy option 8.1: Selection of intermediate periods from the distribution of returns. For calculating the intermediate periods according to this methodology three steps need to be taken. First a number of paths from today to the intermediate period need to be simulated (we require 10,000 simulations for the RHP). For each path to an intermediate point, a value needs to be calculated at the intermediate point by simulating the product to the end of the RHP and discounting back to the intermediate point (again since we require 10,000 simulations the values here would also require 10,000 simulations). Finally the percentiles could be selected at the intermediate holding period. No. Pros Cons 1 Accurate and reliable results. Requires a disproportionate amount of calculations and especially simulations (10,000 * 10,000) to find the values to be displayed at the intermediate periods. Policy option 8.2: Extract from the selected scenarios at RHP the underlying paths A simplified way of calculating the intermediate periods for nonlinear products would be to extract from the distribution of returns at the 10th, 50th and 90th percentile and selecting the paths that lead to the favorable, moderate and unfavorable scenario at the recommend holding period. Then simulate the intermediate values of the PRIIPs on those three selected paths only. No. Pros Cons 1 Feasible to perform for manufactures and supervisors to regulate. Not a suitable methodology for products with caps and knock-outs for example. 2 Meaningful information to investors on more simply structured products Not suitable for more complicated structured products for the information might cause confusion amongst retail investors since the relative performance of the product might be higher for the intermediate periods than at recommended holding period. It could also be the case that for example the moderate scenario could outperform the unfavourable scenario. 3 Questionable whether it provides reliable results for all products in scope.

99
Policy option 8.4: Best estimate for the intermediate periods in the KID No. Pros Cons 1 The information of the intermediate periods shows aspects of downward risk at points before the recommended holding periods. Best estimate might provide too many degrees of freedom to the manufacturer 2 Could provide retail investors with information of the effects of early exit (costs) and the impact on their returns 3 Provides insight on how the product develops over time. 4 The degrees of freedom can be mitigated by general requirements and are further limited by the value of the performance scenarios at the RHP. Policy issue 9: Performance Scenarios presentation format Policy option 9.1: Tables or graphs containing cost information (showing performance both before and after costs) Policy option 9.2: Other formats aimed at showing returns with different levels of likelihood (funnel of doubt, histogram) Policy option 9.3: Tables showing net returns at different holding periods In consumer testing in the first phase five different ways of communicating possible results were explored on objective understanding and comparing and also the subjective preferences of consumers for the different ways of presenting information on performance. From the first phase two variants were found to be both subjectively preferred and objectively understood and well compared by consumers. Those two variants were then again tested within the whole context of the KID. The results of both consumer testing phases were controlled for effects of product and demographical effects, amongst other things financial literacy and experience with PRIIPs products. In the second phase for each successfully tested way of communicating about performance and possible returns, two different variants were tested. A total of four different variants. The second phase did not provide an indisputable preferred option, neither on subjective or objective understanding and comparison. It did show that two of the four variants were outperformed by the other two variants tested in the second phase. Both tables and graphs which show net results for different holding periods (including the recommended holding period) in three different scenarios, were found to be better informative with regards to understanding and comparison and also on what is subjectively the preferred option.
9.1 Tables and/or graphs containing cost information (showing performance both before and
after costs)
No. Pros Cons
1
Reminds investors about the impact of costs on returns Investors have a limited ability to understand/process all the information

100
No. Pros Cons
2 Has a higher informative value Consumers do not subjectively prefer this option. They feel it is too complicated. 3 Consumers have difficulties determining what their possible outcome could be. Probably because the amount of numbers that is presented increases significantly by this option. Also this information is also presented in the cost section.
9.2 Other formats aimed at showing returns with different levels of likelihood (funnel of doubt,
histogram)
No. Pros Cons
1
Some investors appreciate information on likelihood of scenarios A lot of investors have difficulty understanding percentages/likelihood of events 2 When consumers have no information about the likelihood of the scenarios, they automatically make wrongful inferences about the likelihood of the presented scenarios. Therefore, presenting information on likelihood could be preferred. When the aspect of likelihood is introduced this also requires a more prescriptive way of calculating the performance scenarios. This was not the preferred policy option, i.e. at this point it is not possible to introduce the likelihood of the scenarios into the presentation of different scenarios. 3 One of the most important aspects of likelihood, the simple aspect of the fact that the outcomes are uncertain is covered by merely stating multiple scenarios. The exact likelihood of these scenarios occurring is not necessary for this basic principle of understanding possible outcomes.
9.3 Tables showing net returns at different holding periods
No. Pros Cons
1
Simple formats with no additional layers of information are better understood according to consumer testing. However it was also found that only a single timeframe was not providing sufficient information on the possible returns. These formats are less able to show information about the whole distribution of returns, as only a limited number of scenarios can be chosen. 2 Information about returns at different For illiquid products, return information at

101
No. Pros Cons points in time has been found interesting according to the consumer testing. Limiting possible outcomes in order to not confuse them or overload them, but also not depriving them from relevant information on uncertainty of the outcomes. intermediate periods may be misleading. 3 A simpler methodology to select scenarios can be used. Less dependent on a model. The same methodology requirements are needed for a graphical presentation. With regards of to the consumer testing the graphical presentation of the information was sometimes more effective in communicating what possible returns could be expected in a certain scenario at a certain time. 4. In marketing material often graphs are used which makes it hard for the KID to compete for attention in the orientation phase when only tables with performance information are included. Policy issue 10: Treatment of Exchange Traded Derivatives (ETD) Policy option 10.1; ETD’s are provided to have a different treatment for the KID The production of a KID, per the PRIIPS regulatory technical standards developed for other products, and in particular simulations needed for performance scenarios, could require the exchange to update thousands of KIDs several times, potentially every minute, during the course of the trading day. It would also thus be extremely difficult for the correct up-to-date KID to be provided to the retail investor as normally required under the PRIIPs regime. Under this option it is proposed that the performance scenario information is provided in the form of a ‘pay off structure’ graph rather than information on returns to the RHP as for other PRIIPs, and there also would be a specific clarification of the costs to include. The SRI would always by 7 – so the KID could be produced as a result on a generic basis. (Level three measures would be used to help clarify the detailed application to example cases to ensure the KID remains clear, fair and not misleading). The application of a different approach (in respect of performance scenarios and the information on costs in particular), would apply for certain ETDs only – futures, calls and puts – considering these are highly standardized with consistent fundamentals, but which are highly sensitive to the movements on the exchange on an intra-day basis. Further the nature of these derivatives seems to justify a payoff graph being more informative to retail investors, because of their characteristics. For other type of derivatives, for example warrants, turbo’s and CfD’s the information on three different scenarios in which the pay-out is shown is considered more crucial, for often the chances of receiving the pay-out are very slim which is more properly shown in case three scenarios are displayed.

102
Also the standardization, and listing on a regulated market, ensures that this approach would be appropriate from a consumer protection standpoint, but also is a strong reason in view of the practicalities of updating KIDs on an intraday basis in the absence of such an approach. OTC derivatives are by their nature less standardized and less subject to intraday fluctuations across a whole market. For some OTC derivatives transparency and frequency of pricing information is substantial less in comparison with the ETDs. Finally, though OTC derivatives may, of course, also be highly sensitive to fluctuations in underlyings, the lack of standardization and transparency on prices, means it would be difficult to extend the proposed treatment of exchange traded futures calls and puts more widely without raising wider problems. No. Pros Cons 1 It would relieve ETD’s from a disproportional burden to require the exchange to update thousands of KIDs several times. Different treatment could be considered not comparable to other products. 2 Characteristics of the ETDs lead to the fact that for example the performance information is less informative as it is prescribed for other products as it would be presented in a pay-off graph. Derivatives are already relieved from the calculation of the MRM, since these are automatically assigned to risk class 7. 3 ETDs have already highly standardized contract specifications aimed at ensuring that the derivatives are sufficiently liquid to be traded centrally. 4. Although the specific properties can change significantly intraday the fundamental properties remain the same from an investor perspective.
Appendix
Testing the results for the MRM methodology
In order to check whether the MRM is working a selection of products have been tested by calculating the MRM for these products in order to see whether the outcomes would lead to satisfying results. Important to note is that especially a lot of Structured products have been calculated because of their diverging characteristics. By no means this is not intended to be a representative cross Europe sample. It is intended to give an impression of what the consequences are of classifying different PRIIPs according to the MRM methodology.

103
Testing the results for the methodology for performance.
The graph above shows the concentration of values for the different scenarios for each Market Risk class. One might notice from the graph below that the products that are classified in the lower risk classes have a smaller range of outcomes between positive and negative scenarios, considering that the risk and reward tradeoff becomes clear for retail investors. Further the graph shows that indeed the downward risk is increasing following the increase of the MRM, following that lower values are shown for higher risk classes of the MRM. Automatic assignment of PRIIPs without data to MRM 6 The detailed requirements propose an automatic assignment of certain PRIIPs lacking sufficient data to MRM 6. This selection has been on the basis of desk based research, where those PRIIPs with insufficiently monthly data (typically, e.g., those investing in real estate or similar), showed a highest risk class 0 1000 2000 3000 4000 5000 6000 0 1 2 3 4 5 6 7 8 Neg. Perf. Scenario Neutral Perf. Scenario Pos. Perf. Scenario

104 of six. This class contains PRIIPs where there is a very significant chance of losing money, but not due to the very high volatility that might be seen with a derivative or the high risk of a product where you can lose more than your investment. Option 8.1 Tables and/or graphs containing cost information (showing performance both before and after costs) No. Pros Cons 1 Reminds investors about the impact of costs on returns Investors have a limited ability to understand/process all the information 2 Has a higher informative value Consumers do not subjectively prefer this option. They feel it is too complicated. 3 Consumers have difficulties determining what their possible outcome could be. Probably because the amount of numbers that is presented increases significantly with this option. Also this information is already presented in the cost section. Option 8.2 Other formats aimed at showing returns with different levels of likelihood (funnel of doubt, histogram) No. Pros Cons 1 Some investors appreciate information on likelihood of scenarios A lot of investors have difficulty understanding percentages/likelihood of events 2 When consumers have no information about the likelihood of the scenarios, they automatically make their own, wrongful inferences about it. Therefore, presenting information on likelihood could be preferred. Showing likelihood of scenarios requires a more prescriptive way of calculating them. 3 Consumers may attach a disproportionate weight to (some of) the actual likelihood numbers being displayed, rather than concentrating on the overall dispersion of possible outcomes. Option 8.3 Tables showing net returns at different holding periods No. Pros Cons 1 Simple formats with no additional layers of information are better understood according to consumer testing. These formats are less able to show information about the whole distribution of returns, as only a limited number of scenarios can be chosen.

105
No. Pros Cons
2
Information about returns at different points in time has been found interesting according to the consumer testing. Limiting possible outcomes in order to not confuse them or overload them, but also not depriving them from relevant information on uncertainty of the outcomes. For illiquid products, return information at intermediate periods may be misleading. 3 A simpler methodology to select scenarios can be used. Less dependent on a model. The same methodology requirements are needed for a graphical presentation. With regards to the consumer testing the graphical presentation of the information was sometimes more effective in communicating what possible returns could be expected in a certain scenario at a certain time. 4. In marketing material graphs are often used which makes it hard for the KID to compete for attention in the precontractual/reflection phase when only tables with performance information are provided. Policy options for Costs Policy issue1: Investment Funds / Transaction costs Option 1: transaction costs disclosed using a standard table only Option 2: transaction costs disclosed using a hybrid approach, disclosing actual costs where known and a standard table where actual costs are not known The cost disclosure in the KID shall include all types of costs, including both direct and indirect costs. Transaction costs are indirect costs incurred by the PRIIP when it undertakes transactions. The challenge for the KID is that transaction costs are dependent on the transactions that the PRIIP undertakes and so cannot be known in advance. There is also a challenge to ensure that transaction costs, including implicit costs are appropriately calculated. The options considered are: for the ESAs to establish a standard table of transaction costs, and for manufacturers to use this table to calculate the transaction costs within their PRIIP; or for manufacturers to report the actual costs that they incur when they transact, where that is possible, and to use standard costs only when actual costs are not available. After considering the issues and the responses to the Consultation Paper, it was considered that a simple methodology can be defined which enables actual transaction costs to be calculated for

106 most assets, reducing the need for a complex standard table. As of the first annual review ((when there is 6 months of actual data and 30 months of standard table data) the manufacturer might choose to smooth in and use partially actual transactions costs and partially the values in the standardized table." Policy Option 1.1: transaction costs disclosed using a standard table only No. Pros Cons 1 No need to calculate actual transaction costs Calculation is likely to be inaccurate, and may be misleading 2 Potentially easy for industry to calculate estimates of transaction costs Detailed standard table would need to be developed and maintained, using a large number of variables to provide a distinction between higher transaction cost and lower transaction cost assets 3 Manufacturers might not be incentivised to control actual transaction costs as they will not be disclosing them Policy Option 1.2: transaction costs disclosed using a hybrid approach disclosing actual costs where known and a standard table only where actual costs are not known No. Pros Cons 1 Methodology allows simple calculation of actual transaction costs for most assets. In cases where a standard table may be needed would this also be potentially easy to calculate. Standard table still need to be defined and updated 2 Methodology for calculating actual transaction costs is comprehensive, removing investor concerns of hidden costs 3 Uses established methodology already used by some asset managers and supported by independent third parties 4 Investors will be able to compare PRIIPs based on the actual costs they are incurring Policy issue 2: Investment funds / Performance fees Option 2.1: ex-ante estimate of performance fees only shown in performance scenarios

107
Option 2.2: ex-ante estimate of performance fees included in the summary cost indicator (as well as in performance scenarios) Option 2.3: ex-ante estimate of performance fees shown separately in the cost section (excluded from the total cost indicator) The cost disclosure in the KID shall include all types of costs. Performance fees are incidental costs (i.e. they do not appear on a regular basis). They have to be paid under specific conditions, when the performance of the fund is higher than the performance of a predefined benchmark. The challenge that arises for KID as a pre-contractual document is that while the performance fee formula is known in advance of the purchase of the investment, the likelihood and scale of any such fees is uncertain. While it has been chosen to include ex-ante estimate of performance fees in the summary cost indicator, the different options have the following pros and cons:
Policy Option 2.1: only show estimates of performance fees in performance scenarios No. Pros Cons 1 It is potentially easy for industry to integrate performance fees in performance scenarios. Calculation is likely to be error prone, as the choice of values cannot take into account all potential scenarios. 2 - Performance fees would not be indicated as costs: less visibility of these costs for supervisors and investors, as compared to other types of costs. This option may also not be in line with the level 1 PRIIPs Regulation. 3 Investors understand that performance fees impact the reward. Investors will have difficulties to understand that performance fees are costs, as they are not indicated as such. Policy Option 2.2: estimate performance fees ex-ante and include them in the total cost indicator (as well as in performance scenarios) No. Pros Cons 1 Relatively easily to implement for industry. . 2 The total cost indicator would include all potential costs, which would trigger a good visibility for supervisors, and be in line with the level 1 PRIIPs Regulation. The total cost indicator might be error prone because it would show an amount which is higher than the real costs in those years where no performance fee is taken and lower than the real costs in those years

108
No. Pros Cons where the performance is better than the corresponding assumption made for the calculation of the cost indicator. 3 Investors would clearly identify performance fees as costs. Investors might have difficulties to conclude that performance fees will not have to be paid on an ongoing basis (but only under specific conditions). Policy Option 2.3: estimate performance fees ex-ante and show them separately from the total cost indicator No. Pros Cons 1 Relatively easily to implement for industry (as approach is already known for UCITS). Calculation might be error prone in some cases (as based on historical data). 2 Reliable quality of the total cost indication (because errors related to the calculation of performance fees will not be included in the summary cost indicator). The total cost indicator would show an amount which would differ from the real costs This option may also not be in line with the level 1 PRIIPs Regulation. 3 Investors identify performance fees as costs, which are incidental. They might not understand the difference between the summary cost indicator (that would include performance fees) and the incidental costs (performance fees) that could not be simply added to the latter Investors might have difficulties to understand that costs impact the reward. Policy issue 3: PRIPs 12 other than investment funds / Disclosure of certain types of costs for structured products (and derivatives and CFD) Option 3.1: Principle-based approach which establishes a set of high level rules that govern the methodology for the estimation of the fair value of the PRIIP Option 3.2: Prescriptive approach which establishes prescriptive methodologies for the estimation of the fair value of the PRIIP Having in mind that the cost disclosure in the KID shall include all types of costs, the KID of a structured product should enable retail investors to identify, within the issue price of a PRIIP, (i) the amount that will be captured by the manufacturer (cost), and (ii) the amount that will be used to provide/generate a return to the customer (invested capital). To identify the cost components

12 ‘PRIPs’ is used in this CP intentionally to distinguish the smaller collection of non-insurance based products. We follow the definition made in art 4 (1) of the PRIIPs Regulation.

109 that are hidden because included in the purchase price of the PRIIP and discriminate them from the actual invested amount it has been identified the fair value methodology. Alternative approaches which do not use the methodology of the fair value and able to estimate all the costs borne by the investor have not been identified so far. While a principle-based approach to fair value has been chosen, the different identified options have the following pros and cons:
Policy Option 3.1: Principle-based approach to Fair Value No. Pros Cons 1 Ability to define a general formalized process, but which ensures comparability of results Low degree of standardisation 2 Adaption to the increasing complexity of the structure of the products and the evolution of the regulation and the standards

Certainty on the inclusion of all costs in the KID section 4 High level of transparency Policy Option 2: Prescriptive approach to Fair Value No. Pros Cons 1 Standardization Rigidity 2 Difficulties in prescribing a specific model for each type of product 3 Unreliability of the estimations Policy issue 4: Insurance-based investment products / Biometric risk premiums (BRP) and summary cost indicator Option 4.1: Include the costs of BRP (difference between the BRP and the fair value of the BRP) in the summary cost indicator, with no separate figures for BRPs Option 4.2: Exclude BRP (and the costs of BRP) from the summary cost indicator, but show them in a secondary “breakout” The cost disclosure in the KID shall include a summary cost indicator that includes all types of costs of the PRIIP. This summary cost indicator shall be understandable by the investor and allow

110 him to compare the costs of different PRIIPs among one category of PRIIPs, but also the different types of PRIIPs between themselves. While it was chosen to include the costs of BRP (difference between the BRP and the fair value of the BRP) in the summary cost indicator, with no separate figures for BRPs, the different identified options have the following pros and cons:
Policy Option 4.1: Include the costs of BRP (difference between the BRP and the fair value of the BRP) in the summary cost indicator, with no separate figures for BRPs No. Pros Cons 1 Simplest overall picture of all costs This option implies calculating the fair value of the BRP Policy Option 4.2: Exclude BRP (and the costs of BRP) from the summary cost indicator, but show them in a secondary “breakout” No. Pros Cons 1 This option might ease the comparison of insurance-based investment products on BRP related criteria. May be more complex for consumers to understand. This option might not be fully in line with the level 1 PRIIPs Regulation 2 Overall costs of product (combining both BRPs and investment costs) may be less clear Policy issue 5: Insurance-based investment products / Summary indicator for insurance-based investment products Option 5.1: As for the calculation of the summary cost indicator, consider the scenario in which the insured biometric risk does not occur, i.e. only consider premium cash flows and endowment benefit cash flows, both with probability 1. Option 5.2: For the calculation of the summary cost indicator consider all cash flows – premiums, biometric benefit cash flows and endowment benefit cash flows – weighted with the probability of their occurrence. The cost disclosure in the KID shall include a summary cost indicator that includes all types of costs of the PRIIP. This summary cost indicator shall be understandable by the investor and allow him to compare the costs of different PRIIPs among one category of PRIIPs, but also the different types of PRIIPs between themselves.

111
Without a clarification it might not be clear how to treat possible cash flows form insurance coverage when calculation the summary cost indicator. While option 1 was chosen, the different identified options have the following pros and cons:
Policy Option 5.1: For the calculation of the summary cost indicator, consider the scenario in which the insured biometric risk does not occur, i.e. only consider premium cash flows and endowment benefit cash flows, both with probability 1. No. Pros Cons 1 Easy to understand for the investor, yield numbers based on probability weighted cash flows might not be comprehensible for the average investor 2 RIY-approaches following option 1 are already in place in several member states and will therefore be easier to implement for manufacturers 3 Calculations according to option 1 are less complex and therefore less error prone 4 Disregarding probabilities for early termination due to insured events like death is more consistent to other PRIIPsproducts 5 It is not obvious how to treat certain life insurance products without a fix term Policy Option 5.2: For the calculation of the summary cost indicator consider all cash flows – premiums, biometric benefit cash flows and endowment benefit cash flows – weighted with the probability of their occurrence. No. Pros Cons 1 Approach is more general than option 1 2 It is obvious how to apply this approach to life insurance products without fix term 3 Approach is not yet used in practice 4 More complex and therefore more error prone than option 1 5 Yield numbers based on probability weighted cash flows considering benefits

112
No. Pros Cons from insurance coverage difficult to understand for investors Policy issue 6: General / Summary cost indicator Option 6.1: the summary cost indicator is a total cost ratio (TCR) for all types of PRIIPs Option 6.2: the summary cost indicator is a reduction in yield (RIY) ratio for all types of PRIIPs Option 6.3: the summary cost indicator is a RIY, except for those types of PRIIPs where the value of RIY equals TCR and where a TCR is more appropriate The cost disclosure in the KID shall include a summary cost indicator that includes all types of costs of the PRIIP. This summary cost indicator shall be understandable by the investor and allow him to compare the costs of different PRIIPs among one category of PRIIPs, but also the different types of PRIIPs between themselves. While the RIY ratio was chosen for all types of PRIIPs, the 3 options that have been identified include the following pros and cons:
Policy Option 6.1: the summary cost indicator is a total cost ratio (TCR) for all types of PRIIPs No. Pros Cons 1 It is potentially easy to implement a TCR approach for some parts of the industry (funds and structure products). - 2 The TCR approach might be less subject to assumptions on performance Calculation is likely to be error prone for some other parts of the industry (insurancebased PRIIPs), because the timing of the cost deductions is not fully taken into taken by a TCR. 3 Investors will have difficulties to compare the costs of the different types of PRIIPs, because the accuracy of the TCR will be different depending on the type of PRIIP. Policy Option 6.2: the summary cost indicator is a reduction in yield (RIY) ratio for all types of PRIIPs No. Pros Cons 1 It is potentially easy to implement a RIY approach for some parts of the industry (insurance-based PRIIP)

113
No. Pros Cons
2
The RIY approach is equally accurate and applicable for all types of PRIIPs. The RIY approach might be more subject to assumptions on performance. The implementation of the RIY approach to structured products needs to be more specified than for the other types of PRIIPs. 3 Investors will be able to compare the costs of the different types of PRIIPs. Policy Option 6.3: the summary cost indicator is a RIY, except for those types of PRIIPs where the value of RIY equals TCR and where a TCR is more appropriate No. Pros Cons 1 It is potentially easy to implement for all parts of the industry (funds, structured products and insurance-based PRIIP) 2 This option would mean using a different methodology to calculate the summary cost indicator for the different types of PRIIPs, which might raise issues of consistency3 Investors might have difficulties to understand that the value of the RIY equals the value of the TCR for some types of PRIIPs and why for some products a TCR approach is shown while for some others a RIY approach is shown. Policy issue 7: Presentation format of the cost section of the KID Option 7.1: Keep both detailed tables (showing costs over time and the break-down of costs) as included in the consultation paper Option 7.2: Shorten the size of the table on costs over time in order to simplify the overall information shown to the consumer The cost disclosure in the KID shall include a summary cost indicator that includes all types of costs of the PRIIP, and that show costs over time. The cost disclosure shall also show a breakdown of costs. The amount of information shown in the cost section of the KID shall be such that the information is comprehensive, but shall not be such that the investor is not incentivized to fully read it carefully. While the option to narrow down the table on costs over time was chosen, the 2 options that have been identified include the following pros and cons:

114
Option 7.1: Keep both detailed tables (showing costs over time and the break-down of costs) as included in the consultation paper No. Pros Cons 1 The cost section of the KID would include information on the different sums of the different types of costs (one-off costs, ongoing and incidental costs). 2 The information included in the cost section of the KID might be such that the investor would not incentivized to fully read it carefully. Option 7.2: Shorten the size of the table on costs over time in order to simplify the overall information shown to the consumer No. Pros Cons 1 The information included in the cost section of the KID would be such that the investor is more incentivized to fully read it carefully, as compared to option 7.1 2 The cost section of the KID would include less information on the different sums of the different types of costs (one-off costs, ongoing and incidental costs).
8. Analysis of overall impacts
The assessment of the ESAs, following consultation through the relevant Discussion Papers, has been to retain the options highlighted above. In summary:
Issue Retained option
Risk 1 Selection of types of risk to include in the SRI:
Focus on market, credit and liquidity risk
Risk 2 Aggregation of main types of risks
Aggregation of market and credit, with liquidity risk shown in warning in indicator Risk 3 The different aspects of the Methodology to get to a Summary Risk Indicator Mixture of quantitative measure with qualitative elements Risk 4 Disclosure of liquidity risk and liquidity profile (liquidity issues) Warning within indicator Risk 5 Selection of SRI time frame End of maturity or Recommended Holding Period

115
Risk 6 SRI presentation format
Simple graphic presentation of scale
Reward 1 Choice of Performance Scenarios
Flexible approach
Reward 2 Performance Scenarios presentation format Tables, without indication of likelihood Costs 1 Investment Funds / Transaction costs Hybrid approach Costs 2 Investment funds / Performance fees Ex-ante estimate included in performance scenarios and costs Costs 3 PRIPs other than investment funds / Disclosure of certain type of costs for structured products (and derivative and CFD) Fair value methodology, common principles Costs 4 Insurance-based investment products / Biometric risk premiums (BRP) and summary cost indicator RIY Costs 5 Insurance-based investment products / Summary indicator for insurance-based investment products Calculation assuming biometric risk does not occur Costs 6 General / Summary cost indicator RIY comparators Overall Assessment In the view of the ESAs, the retained options provide for an optimal balance when considered in the context of the criteria identified above, and in achieving the overall objective of consistency in approaches to ensuring the comparability and comprehensibility of the KID. A high level of standardisation and comparability is ensured, coupled with a sufficient and appropriate degree of flexibility given the wide range of PRIIPs in scope, as necessary for ensuring KIDs that are accurate for specific products. Costs for the industry, including for smaller PRIIP manufacturers, have, where possible, been taken into account. Impacts for supervisors have also been considered. Crucially, in view of the goal of the KID, all of the retained options have been considered so as to take into account the results from the Consumer Testing, so as to optimise the approach in view of the use of the KID by consumers, and thereby the comprehensibility and comparability of different options in practice for a representative sample of retail investors. Specific Costs Both one-off and ongoing costs are entailed by the introduction of the KID. As noted above, to a large degree both of these types of costs are entailed by the requirements already set at level one.

116
In regards the use of quantitative methodologies as part of the calculation of the SRI, one off costs can be anticipated for many PRIIP manufacturers in identifying and developing the expertise, processes and data sets necessary. On-going costs can be anticipated in relation to monitoring, continued data acquisition, and reviewing and republishing the KID. These costs can be expected to be similar to those experienced by UCITS when introducing the SRRI for these funds, though some additional aspects necessary for the PRIIPs SRI may increase one off costs to a degree for all PRIIP manufacturers. Economies of scale are likely to reduce these costs for larger PRIIP manufacturers, while it can be expected that third party support for smaller PRIIP manufacturers will also reduce the impact. In regards the cost methodologies, the major change for funds will be the need to now develop processes for calculating portfolio transaction costs. Views amongst stakeholders vary on the scale of these costs, and it can be expected that ongoing costs will drop once systems are in place and once necessary data has been identified. The calculation of RIY figures as opposed to TERbased ongoing cost figures, as are now common for funds, should incur relatively low costs in practice (an RIY can largely be calculated from the TER, once portfolio transaction costs are included and one-off costs are estimated). For structured products, the ‘fair value’ estimation will incur one-off costs related to setting up procedures and data availability, and ongoing costs for monitoring, data collection, KID republication. Costs may be reduced given existing pricing and risk management methodologies implied internally by the relevant PRIIP manufacturers. For insurance-based investment products, RIY methodologies exist in some markets, but there will be one-off costs in particular in relation to the costs associated with biometric risk premiums, where new processes and data will need to be developed. These costs may be mitigated in view of data necessary for internal processes already in place, including in regards Solvency II. MOPs Costs The costs related to the KID for multi-option products are worth highlighting, though these costs are in very large part triggered by the level one. In some markets unit-linked insurance contracts have been offered where disclosures related to units offered in these contracts have been provided by means of UCITS KII documents. These UCITS are used to back units offered in the insurance contract. In this context there are two important consequences that may be foreseen under the PRIIPs Regulation:
 The Regulation provides a temporary exemption whereby UCITS continue to provide KII, and are not required to produce KIDs. However, the insurance undertaking offering a unit-linked insurance contract is themselves a PRIIP manufacturer in the meaning of the PRIIP Regulation, and liable to produce a KID for the unit-linked contract, including either in this KID or in separate documents detailed information in respect of the investment options offered. The combination of the general KID for the unit-linked contract and the more detailed documents in respect of the investment options offered must comply with
Article 8 (3) of the PRIIPs Regulation. This creates a problem because the requirements in

117 the Regulation are not consistent with the information produced in a KII (the risk indicator and cost disclosures, for instance, are different). In so far as the insurance undertaking requests that the UCITS to already prepare KID compliant information for the insurance undertaking – though this is not directly required under the PRIIP Regulation -- this would have the practical impact of undermining the temporary exemption in the Regulation and requiring UCITS to produce simultaneously a KII and KID-compliant information.  More generally, in so far as the information provided about the unit-linked contract must comply with Article 8 (3) of the PRIIPs Regulation, the information provided about the investment options must take into account the difference between the investment option offered and the instruments used to back that option. For instance, the costs and performance of the investment option may not be identical to the situation if the instrument to back it was bought directly by the retail investor, while the legal form of the investment would be different. The PRIIPs Regulation applies at the level of the PRIIP Manufacturer (the MOP provider) and does not apply obligations on the providers of underlying investments. A PRIIP manufacturer who offers a PRIIP that invests in other PRIIPs (whether in the form of a MOP, or otherwise, as in the case for instance of a fund-of-funds) would be able to obtain all the information necessary for the preparation of the disclosures they are responsible for on a look through basis from the published KID for that other PRIIP, and should not require additional information. This should limit indirect impacts where a KID for an underlying investment is available. Lack of Grandfathering The level one does not foresee any grandfathering. Given the volume of individual PRIIPs outstanding currently, including being traded on secondary markets, there could be very significant day one compliance and cost impacts. PRIIPs traded on exchanges There will be new one-off and ongoing costs for PRIIP manufacturers who hitherto have not had equivalent pre-contractual disclosure requirements in regards to transactions on these markets. These costs could be different depending on whether the PRIIP was first issued in a primary market – for which a KID would also be prepared – or whether the PRIIP is only been available on exchange. For collective investment schemes distributed via secondary markets the impact would be less, as these will already normally be required to produce pre-contractual disclosures also for transactions on the secondary markets. The fact that these costs arise follows the level one requirements. The information necessary for PRIIPs traded on exchanges has been considered also in view of obligations related to the update and revision of the KID. In this regard, the information required in the PRIIPs has been calibrated so as to avoid a situation in which continuous (‘real time’) updating is
needed. The KID is conceived as a pre-contractual document, rather than a ‘contract note’ or similar specific confirmation or statement of a proposed transaction. In this view the information within the KID has been carefully developed to ensure the KID can have general applicability and does not need to be updated on a continuous basis.

118
Benefits
The KID is introduced to address consumer issues in the retail investment markets – poor decision making, difficulties understanding and comparing products. The impact of these issues – reduced confidence and trust in financial markets and services, on the one hand, and individual detriment (whether in terms of absolute losses or opportunity costs) – may be significant. The level one impact assessment was not able to fully quantify the possible impact of improving disclosures in monetary terms, given the range of factors that determine outcomes for consumers, including the impact of problems in advice processes, but did note that the massive scale of the EU investment market means even small impacts (for instance, a less than 1% reduction in mis-selling) can outbalance costs. The measures in the draft RTS are designed to take into account consumer testing results, precisely to try to maximise their effectiveness in practice for retail investors, in view of both the comparability and comprehensibility of KIDs, but also in view of the ability of these documents to ‘engage’ the retail investor and encourage their use of the document. Improved transparency can also be beneficial indirectly through its impact on other market players than the retail investor themselves. For instance, greater clarity on costs, risks and performance is likely to aid advisors and distributors when considering which investment proposition to recommend to a customer. As mentioned above in the discussion of the baseline, the establishment of standardised methods for the production of the KID – while these of course incur costs – could also reduce these costs for the industry compared to the situation where the level one requirements are not further specified at level two. Legal uncertainty and the costs of establishing – case by case – an approach that is sufficient for satisfying the level one requirements could incur significant costs, and by contrast, standardisation can be expected to reduce legal uncertainty and associated risks – including compliance risks -- and reduce one-off costs associated with scoping out and implementing PRIIP manufacturer specific policies. Standardisation will also have benefits for supervisors by reducing implementation costs at the supervisory level. In cross-border situations standardisation will also benefit supervisors, PRIIP manufacturers and distributors, and retail investors, by reducing the scope for misunderstandings and barriers to such activity. Consistency will aid in building the confidence of those buying cross-border, and in ensuring they clearly understand their rights and the nature of any commitment they enter into. Other impacts The retained options may have some differential impacts for different stakeholders. In general terms, smaller PRIIP manufacturers may face costs that are not proportionate to their size, where they are not able to draw on internal expertise or resources or economies of scale. Further to this, while
KID requirements will incur one-off and ongoing costs for all PRIIP manufacturers, those that already have in place retail disclosure processes can be expected to be

119 bear lower costs that those setting up such processes for the first time. This however is clearly a consequence of the level one requirements. There is also a possibility – given the heterogeneity of the PRIIPs market across the EU – that differentials in cost impacts cluster geographically. However, as mentioned, these impacts are driven in very large part by the level one requirements, rather than the specific measures foreseen at level two. Any cost impacts for PRIIP manufacturers or those indirectly impacted as providers of underlying investments used by PRIIP manufacturers could in theory reduce the availability of PRIIPs, while enhanced transparency can be expected to have some competition impacts with similar consequences for the availability of PRIIPs across the EU. It is however very difficult to reliably assess such possible impacts. The consequences for retail investors of such factors – for instance, reduced prices or changes in access (either a reduction in the range of PRIIPs offered, or an increase in the range) -- are equally difficult to assess.
6. Monitoring indicators
The main objective for these RTS is consistent application of Article 8 (3), and in particular consistency in measures to improve comparability and comprehensibility of risks, rewards and costs for PRIIPs. For risks, a ‘SRI’ has been introduced which aggregates market and credit risks; for rewards performance scenarios are shown which use the market risk measure as a basis; for costs, summary indicators calculated in a consistent way are required. General monitoring indicators that have been identified are:
 Reduction (over longer term) in successful complaints related to pre-contractual information and PRIIPs in general  Increased confidence (over longer term) amongst retail investors, including in relation to cross-border business  More competitive markets for different types of PRIIPs across EU More specifically for the detailed measures required under these RTS:
 SRI testing across wide range of products shows accurate and appropriate rankings, e.g. in view of independent experts  Limitations of performance scenarios effectively understood by customers, and forecasts are appropriate, e.g. in view of independent experts  Cost information used by retail investors for comparisons, including in the context of third party comparison tools

120
4.1.2 Impact Assessments for RTS under Article 10 (2)

Procedural issues and consultation of interested parties
According to ESAs’ Regulation, the ESAs conduct analysis of costs and benefits when drafting regulatory technical standards. The analysis of costs and benefits is undertaken according to an Impact Assessment methodology. The draft RTS and its impact assessment are subject to public consultation. In preparing these draft RTS, the ESAs have conducted two separate public consultations by means of Discussion Papers.
Problem definition
According to Article 10, the ESAs shall develop draft regulatory technical standards specifying:
A. Conditions for review of the information contained in a key information document (KID); B. Conditions under which the KID must be revised;
C. Specific conditions for review and revision for a PRIIP ‘made available to retail investors
in a non-continuous manner’; and
D. Circumstances in which retail investors are to be informed about a revised KID and the means for this. The general purpose of further specifying these conditions and circumstances is set out in the Recitals to the PRIIPs Regulation, where it is stated in Recital 21 that ‘detailed rules’ are necessary ‘to ensure that the information contained in the key information document is reliable’. Once a KID has been prepared and published, it is possible that developments in the financial markets, amongst other changes, could lead to a situation in which the information contained in a KID is no longer sufficiently accurate as to give a correct basis for an investment decision in relation to the PRIIP it relates to. Information on the potential risks, performance, and costs of the PRIIP can be expected to be most liable to change. For some PRIIPs that continue to be sold to investors, the costs will indeed vary from year to year. For this reason, it shall be necessary to assess the continued accuracy of the information contained in the KID – pursuant to the methodologies to be followed for the risk, performance and cost disclosures as set out above in regards to the RTS under Article 8 – so that the KID that is made available to retail investors can be used as a basis for an informed investment decision. However, while some quantitative factors on which information in the KID is dependent may change on a continuous basis, it is not proportionate nor relevant, given the summary nature of the KID and the methodologies to be followed in preparing the risk, performance and cost information set out in the RTS under Article 8, for the KID to be produced on a ‘real time’ basis, notwithstanding the fact that its content might have to be monitored on a close or continuous basis depending on the nature of the PRIIP. For instance, a certain degree of averaging or

121 smoothing of data, and use of approximations, is foreseen in the methodologies using quantitative information on costs and for the summary risk indicator. On this basis, therefore, questions arise as to the methods to be followed for reviewing the changing circumstances, in view of their possible impact on the information in the key information document, in revising the key information document where necessary, and in ensuring it is made available to prospective retail investors. It can also be considered that changes may occur that are sufficiently important so as to enable retail investors who are already invested in a PRIIP to be informed of these changes, as these changes may be material for a decision by those investors as to whether they should remain invested in the PRIIP. In the absence of sufficient minimum standards on these methods to be followed for review, revision, and republication, it can be expected that in some cases retail investors will receive key information documents that are not sufficiently accurate. This could then lead to mis-buying of PRIIPs, with consequent possible consumer detriment in terms of opportunity costs for retail investors (asset allocations that are not efficient for those retail investors) or outright losses for those investors (for instance, by those investors taking on unforeseen risks that are, because they are unforeseen, not effectively mitigated). Baseline scenario When analysing the impact from proposed policies, the impact assessment methodology foresees that a baseline scenario is applied as the basis for comparing policy options. This helps to identify the incremental impact of each policy option considered. The aim of the baseline scenario is to explain how the current situation would evolve without additional regulatory intervention. The baseline scenario is based on the situation where the PRIIPs Regulation applies, but where there is no RTS to further specify the obligation in Article 10 (which states under 10 (1) that the PRIIP manufacturer shall ‘review the information contained in the key information document regularly and shall revise the document where the review indicates that changes need to be made. The revised version shall be made available promptly’.)
3. Objectives
The overall objective of this RTS is to ensure consistency in the application of measures to be taken by PRIIP manufacturers to ensure that retail investors do not make investment decisions on the basis of KIDs that have become inaccurate or misleading following publication. Such consistency would ensure that retail investors can be confident that the KID available to them can be relied on – whoever the PRIIP manufacturer or distributor is, and irrespective of their jurisdiction. The assessment of the ESAs is that the specific policy issues to be addressed, in view of the scope of the empowerment for draft RTS, are:

122
1: The processes and procedures to be put in place by PRIIP manufacturers 2: Timing of periodic reviews 3: Criteria for republication of the KID, including materiality of change criteria 4: The treatment of non-continuous offers 5: Communication with retail investors
4. Policy options
Policy issue 1:The processes and procedures to be put in place by PRIIP manufacturers Option 1.1: Highest degree of flexibility for manufacturer Option 1.2: Common principles (minimum standards) Option 1.3: Detailed rules In order for the PRIIP manufacturer to be in a position to be able to ensure a published KID continues to be accurate, that manufacturer would need to be able to identify and monitor circumstances whereby the information in the KID may be no longer accurate. A variety of circumstances might have such an impact:
 Fundamental changes to the product itself, such as changes to its investment strategy or objectives  Changes in the pricing of underlying assets or reference values, or to their risk/reward profile, that could lead to a change in the product’s risk rating  Changes in the pricing of underlying assets or reference values or to their risk/reward profile, that could impact the product’s performance information  Costs that impact the investor arising that were not foreseen, or where the volume or level of currently disclosed costs is different This is of particular significance where changes are occurring outside of a normal review cycle. Given the range of PRIIPs and PRIIP manufacturers within the scope of the Regulation, the types of circumstance that might need to be considered will be very wide, and it would be difficult to identify all such circumstances. In addition to the identification of kinds of circumstance as a one-off activity, the PRIIP manufacturer would need to be in a position whereby they can monitor these circumstances. Some of these circumstances would be clearly under the control of the manufacturer (such as changes of strategy, or a change to the fee structure or level), such that any need to review and republish a KID could form a natural part of the manufacturer’s activities in relation to that circumstance. Other circumstances are liable to change continuously (such as those related to the pricing or valuation of assets). In these cases, the appropriate frequency and accuracy of any monitoring processes would need to be considered. What would be appropriate would depend also on the

123 question of what amounts to a ‘material’ change such that it should trigger a change in the contents of the KID and hence a republication of the KID. This relates to policy issue 3 examined below, Again, the range of PRIIPs and the circumstances that may impact the accuracy of the KIDs produced for these PRIIPs means that the processes effective for identifying and monitoring these circumstances are likely to vary significantly. For instance, for an open ended fund, it may be important to set a process for monitoring and responding to movements in the pricing of underlying assets if defined risk triggers are exceeded, or to set a process for monitoring portfolio transaction costs. For a fixed term structured instrument, the pay-out formula and fees may typically be predetermined, but substantive changes in the behaviour of underlying assets or reference values could nonetheless trigger a change in the instrument’s risk rating, so processes would be needed for monitoring their behaviour. Insurance based investment products could see changes in the outlook for the insurer that impact likely bonuses to a degree that causes a change in their risk rating, again such that monitoring and the use of triggers would make sense. Policy Option 1.1: Flexibility for manufacturers No. Pros Cons 1 Manufacturers able to achieve maximum proportionality to their specific products, business models and operating scale and structure. Approaches of different manufacturers diverge, reducing effectiveness of the KID for comparisons. 2 - Approaches of different manufacturers diverge, increasing risk of misleading KIDs for retail investors. 3 - Some legal uncertainty/legal risk for manufacturers in assessing compliance. Policy Option 1.2: Common principles (minimum standards) No. Pros Cons 1 Greater legal certainty for manufacturers than option 1.1. Minimal standards may still leave room for divergence in approaches, with risk of misleading and less comparable KIDs. 2 Minimum standards ensure common minimum level of comparability and accuracy of KID for retail investors. Minimal standards may still leave room for some legal uncertainty/legal risk for manufacturers in assessing compliance. 3 Flexibility for manufacturers to adapt procedures for their products and organizational structure – proportionality in costs. Policy Option 1.3: Detailed rules

124
No. Pros Cons
1 Greatest legal certainty for manufacturers.
Detailed rules may not be always appropriate for every PRIIP or manufacturer, so some scope for gaps / risks that are not covered. 2 Comparability and accuracy of KID for retail investors. Lack of flexibility for manufacturers to adapt procedures for their products and organizational structure – disproportionate costs for some. Policy issue 2: Timing of periodic reviews Option 2.1: Left for manufacturers to decide Option 2.2: Yearly reviews in all cases Option 2.3: Minimum standard for reviews (at least yearly) The timing of periodic reviews is a specific element of the common standards addressed already under Policy issue 1 that merits individual treatment. In general, periodic reviews are necessary for certain parts of the information in the KID. Typically the information on costs, risks and performance can be expected to be determined in such a way that it needs to be ‘updated’ after each year. This is the approach that is already used for the purposes of the UCITS KII. This would immediately imply a fixed periodic review on a yearly basis, supplemented by ad hoc reviews where circumstances have changed so as to indicate that the KID may be inaccurate. However, given the range of PRIIPs and PRIIP manufacturers, such a period for the review may not always be so relevant – for instance, some fixed maturity PRIIPs may not have costs that are varying on an ongoing basis, or may have shorter lifespans, accordingly meriting a shorter frequency of periodic review. An alternative, in view of the range of PRIIPs, would be to leave the choice of period to the manufacturer, in view of their overall liability for the accuracy of the KID. The frequency of review could significantly increase costs – where reviews are conducted on a frequent basis, such as on a quarterly basis, for some PRIIP manufactures who issue large numbers of PRIIPs, particularly if linked to mandatory revision and republication of the KID. This may seem disproportionate where there are only small changes that for instance do not lead to a different cost, performance or risk disclosure. On the other hand, as long as revision implies only publication on the website of the manufacturer, publication of the revised version through yearly mandatory reviews may not lead to significant additional costs, while ensuring a minimum update frequency of the information in all cases.

125
Policy Option 2.1: Left for manufacturers to decide No. Pros Cons 1 Manufacturers able to achieve maximum proportionality to their specific products, business models and operating scale and structure. Approaches of different manufacturers diverge, reducing effectiveness of the KID for comparisons. 2 - Approaches of different manufacturers diverge, increasing risk of misleading KIDs for retail investors. 3 - Some legal uncertainty/legal risk for manufacturers in assessing compliance. Policy Option 2.2: Yearly reviews No. Pros Cons 1 Model has worked well for the UCITS market, and likely to align well with costs, risk and performance review cycle (Art 8). Yearly cycle may be either too infrequent or frequent for some PRIIPs: risk for retail investors, and for manufacturers. 2 Legal certainty. Policy Option 2.3: Minimum standard for reviews (at least yearly) No. Pros Cons 1 Comparability and accuracy of KID for retail investors – KID reviewed more frequently where necessary. Some legal uncertainty over choice of periodicity. 2 Adaptability for shorter term PRIIPs or those where a yearly cycle is too short.

126
Policy issue 3: Criteria for republication of the KID Option 3.1: Wide scope: following all reviews Option 3.2: Narrow scope: where review shows need for changes Once a review has been undertaken, a decision needs to be taken as to how the KID will be revised and thereafter republished. This relates to the important materiality test for changes. As set out above, a republication may be envisaged after every periodic review. This would be to follow the approach in the UCITS rules related to KII documents. There, the methods followed for the cost information, the risk indicator, and the past performance information contained in the document, imply that the KII will need to be republished at the very least on a yearly basis. The cost methods outlined in this Consultation Paper foresee the use of ex-post cost data on a yearly basis, in a similar fashion to the KII for UCITS, and for this reason a similar republication cycle can be foreseen to take into account the most up to date data. For ad hoc reviews it may be very often the case that following the review a revision and subsequent republication is required, since the ad hoc review would typically only be triggered where a new KID would likely be necessary. For both periodic and ad hoc reviews, however, the question arises as to the degree of ‘materiality’ of changes – for instance, a change of a basis point or two in costs may not lead to a change in the costs figures shown in the KID, once rounding and approximation is taken into account. While the KID might be republished as normal, the figures and contents of the document would remain the same. The cost methodology foresees recalculation on a yearly basis to reflect changes that impact the summary cost indicator at the level of 2 decimal points. Where there are changes to the costs during any year that are known to the PRIIP manufacturer, where such changes would be material for the retail investor, these would also trigger an ad hoc revision and republication. In regards the risk methodology, under the quantitative measurement of market risk, a PRIIP might temporarily move between SRI categories, or, where it sits close to the boundary between buckets, the assigned SRI may be particularly sensitive to small changes in the risk outlook. The methodology foreseen for the risk indicator includes some intrinsic smoothing and approximations that would to a degree reduce excessive sensitivity. An additional ‘smoothing’ is foreseen however within the risk methodology, whereby a movement between SRI categories related to market risk must persist for at least four months before it is to be counted as a permanent change triggering a revision and republication of the KID. Shorter term movements would not require republication. Changes related to credit risk would however always require revision and republication of the KID. In the absence of these smoothing and approximation measures, a clearly disproportionate approach could arise where the KID would
need to be continuously updated and republished; such an approach was not considered a viable option.

127
A choice remains however as to whether the KID would be always republished following a review, or whether the KID might only be revised and republished following changes. Where the KID is always revised and republished, there could be benefits from the legal certainty this brings for issuers, the market and retail investors, in that to find the accurate KID, one would need to simply locate the KID closest to the nearest periodic review date. Where the revision and republication are left to manufacturers to decide, questions on the materiality of changes – and whether these should lead to revision and republication – might arise, while different practices emerge in different markets and sectors. Policy Option 3.1: Wide scope No. Pros Cons 1 Maximises accuracy of KID . Could imply high costs. 2 Legal clarity. Costs may be disproportionate for some manufacturers. Policy Option 3.2: Narrow scope No. Pros Cons 1 Lower costs for some manufacturers Legal uncertainty. Policy issue 4: The treatment of non-continuous offers Option 4.1: Apply same rules as for continuous offers Option 4.2: Apply rules solely where a PRIIP is available – or soon to be available – to retail investors The case of non-continuous offers is raised specifically in level one. These occur where a product is offered to retail investors for only a pre-defined period. Once that period is complete, the product may no longer be offered to retail investors.

128
Note that, in general, periodic and ad hoc reviews are relevant where a PRIIP remains ‘open’ to retail investors. In principle, if the product is no longer on offer to new investors, there will be no new investment decisions to be made by such investors. This leaves solely investment decisions (related, e.g. to opportunity costs) for existing investors. An alternative possibility, where a product is available in a discontinuous fashion, could be a product that is only available during specific ‘windows’. In this regard, there is a risk to retail investors if the PRIIP is not reviewed and republished prior to it being made available (again), so as to ensure its continued accuracy. Policy Option 4.1: Apply same rules as for continuous offers No. Pros Cons 1 Simplicity of rules, applying evenly to all PRIIPs. Level one empowerment foresees specific treatment. 2 Revision of document whilst a PRIIP is not on sale may be disproportionate. Policy Option 4.2: Apply rules solely where a PRIIP is available – or soon to be available – to retail investors No. Pros Cons 1 Proportionate costs, where a PRIIP is not on offer. More complexity in applying rules. 2 Coherence with level one. Policy issue 5: Communication with retail investors Option 5.1: Active model Option 5.2: Passive model Where there is a change to the KID, the new KID should be used by manufacturers and distributors when they are selling directly for satisfying requirements to provide the KID to retail investors.

129
However, there may be cases where existing investors should be aware also of the changes in the KID. A shift in the risk indicator or costs of the PRIIP may be material for the existing investors in their decision as to whether to remain invested in the PRIIP in question or to shift their investment to another PRIIP. Two models can be envisaged – a passive model, where the KID is republished, but it is up to individual investors to monitor the information and locate and use the new KID, and an active model, where the KID is provided directly to investors (e.g. through email, through physical paper mailings, or otherwise). For UCITS, a mostly passive model is followed, though some structural changes to UCITS (mergers, for instance) may trigger a requirement to actively provide a KID in respect of that structural change. A challenge for an active model is that the PRIIP manufacturer may not have access to the identity of their underlying retail investors. In such a case an active model would need to address the entire distribution chain, and obligations solely on the PRIIP manufacturer would be unlikely to achieve the desired impact. It could be very costly therefore to put in place a requirement for active communication, in that this could require the creation of single registers of retail investors for all PRIIPs, irrespective of their legal form. Such measures would impact some PRIIP manufacturers to a far greater extent than others. Policy Option 5.1: Active model No. Pros Cons 1 Investors will be systematically informed of evolution of their investment. Likely to be costly to implement.. 2 May cause significant market disruption (in order to comply). Policy Option 5.2: passive model No. Pros Cons 1 Cheap to implement. Some investors will not receive information or appropriately monitor their investment. 2 Investors may still gain access via website (tools may be implemented to enable ‘alerts’ to those investors interested in information).

130
5. Analysis of impacts
Rules on review, revision and republication of the KID can be significant cost drivers for PRIIP manufacturers, yet are important for ensuring retail investors can be confident that the information they are provided in a KID can always be relied on. The absence of coordinated rules in this area would not necessarily reduce costs, as PRIIP manufacturers are bound to ensure the KID used for all sales of a PRIIP is fair, clear and not misleading, and so could be expected to establish internal procedures and governance for these to ensure timely review and republication of the KID in those situations where it may no longer be fair clear and not misleading. Cost drivers that have been identified include one-off costs related to introducing procedures and measures to ensure compliance, and the ongoing costs of operating these procedures. The costs can be expected to be split between those that are sensitive to the number and frequency of KIDs issued by a manufacturer, and the frequency of review necessary for these PRIIPs, and those that are fixed. One-off costs could include consultancy fees, internal resources for establishing policies and putting these into place. Ongoing costs could include publication costs (updates to websites, provision to distribution channels of new KIDs, and ensuring old KIDs are no longer in circulation or use). The ESAs analysis is that a significant factor in the scale of these costs would be the frequency of reviews and revisions, and the extent or use of an ‘active’ communication model for existing retail investors. The latter could be by far the greatest cost driver, in that in the absence of a relationship between the manufacturer and retail investors, new mechanisms would need to be put in place. The cost impact of this Regulation is significantly reduced, in view of the proposal to follow a ‘passive’ model for communication with retail investors, and the proposal to normalise on a 12- month period for periodic reviews. The expectation is that the costs borne by standardisation of measures on periodic review, revision and republication, including by setting a standardised maximum period for republication (annual), are likely to be materially similar to those borne by virtue of the level one Regulation and its high-level obligation in relation to review, revision and republication of the KID. Such additional specification as is imposed by means of this Regulation is, in the view of the ESAs, unlikely to drive significant additional costs. In practice, this specification could reduce one-off costs in particular, by reducing legal uncertainty and compliance an implementation costs by virtue of the employment of external consultants. Benefits The harmonisation of measures on review, revision and republication of the KID can be expected to ensure greater consistency in the practices adopted by PRIIP manufacturers, and thereby to contribute to increased confidence by retail investors when using the KID. Greater reliance on
information in the KID could have significant impacts. In addition, in so far as the greater harmonisation of this Regulation reduces cases in which retail investors make decisions on the

131 basis of information that is not fair, clear and not misleading, then this would directly contribute to their benefit. For firms, clearer rules would reduce compliance costs.
6. Monitoring indicators
The overall objective of this RTS is to ensure consistency in the application of measures to be taken by PRIIP manufacturers to ensure that retail investors do not make investment decisions on the basis of KIDs that have become inaccurate or misleading following publication. Such consistency would ensure that retail investors can be confident that the KID available to them can be relied on – whoever the PRIIP manufacturer or distributor is, and irrespective of their jurisdiction. Monitoring indicators that have been identified are:
 Availability of up-to-date KIDs for different PRIIPs and distribution channels  KIDs updated in timely fashion following changes in SRI, cost structures  Consistency in outcomes across different PRIIP manufacturers for identifying relevant changes and updating KIDs

132
4.1.3 Impact Assessment for RTS under Article 13 (5)

Procedural issues and consultation of interested parties
According to ESAs’ Regulation, the ESAs conduct analysis of costs and benefits when drafting regulatory technical standards. The analysis of costs and benefits is undertaken according to an Impact Assessment methodology. The draft RTS and its impact assessment are subject to public consultation. In preparing these draft RTS, the ESAs have conducted two separate public consultations by means of Discussion Papers.
Problem definition
According to Article 13 (5), the ESAs shall develop draft regulatory technical standards specifying the conditions for fulfilling the requirement to provide the KID in ‘good time before’ the retail investor is bound by any contract or offer relating to the PRIIP. The general purpose of further specifying these conditions is to ensure consistent application of
Article 13 (1).
The KID has been designed as a pre-contractual document, to aid the retail investor in understanding and comparing different PRIIPs. The success of the KID in this regard depends therefore on the KID being read prior to any investment decision being taken by the retail investor. This means the KID must be provided sufficiently early in the deliberation process of the retail investor as to be able to assist that process. The KID is not designed to inform the retail investors about the product they have bought, or to confirm to the retail investors the commitment they have made. The KID is also a response to a problem faced by retail investors in understanding and comparing retail investments, due to challenges related to the complexity of these investments and investment decisions in general, and relatively low levels of engagement and financial literacy amongst retail investors in general. The timing of the delivery of the KID so as to allow the retail investor to reflect on its contents therefore needs to take into account the problems faced by investors. In addition, neither retail investors nor PRIIPs they are considering are always the same in terms of the amount of time needed for the retail investor to comprehend the PRIIP. The comprehension of the PRIIP by means of the KID would depend on the complexity of the PRIIP itself, the novelty of that PRIIP for the specific retail investor, and the financial capabilities of that retail investor. While the level one Regulation sets out a common principle on the timing of delivery (‘in good time’) the interpretation of this for different types of PRIIPs and different types of retail investors is not specified. For distributors this may raise legal concerns over their liability in relation to the timing of their delivery of the KID for a PRIIP. This could include, for instance, questions over whether or when a prospective retail investor might be expected to be given the time to take KIDs

133 for proposed products away from the distributor, for instance to engage in further deliberation prior to making a commitment. The main problems to be addressed therefore are:
 Clarity for PRIIP distributors over regulatory / legal expectations  Responding to the different needs of retail investors  Responding to the different PRIIPs on offer Baseline scenario When analysing the impact from proposed policies, the impact assessment methodology foresees that a baseline scenario is applied as the basis for comparing policy options. This helps to identify the incremental impact of each policy option considered. The aim of the baseline scenario is to explain how the current situation would evolve without additional regulatory intervention. The baseline scenario is based on the situation where the PRIIPs Regulation applies, but where there is no RTS to further specify the obligation in Article 13 (which states under 13 (1) that ´A person advising on, or selling, a PRIIP shall provide retail investors with the key information document in good time before those retail investors are bound by any contract or offer relating to that PRIIP.’) Note that the publication requirement placed on the PRIIP manufacturer under Article 5 (1) already addresses an aspect of the problem of retail investor access to the KID during the deliberative phase of an investment decision process: this publication obligation will ensure the generalised availability of the KID for those retail investors who are willing and active to find further information during an exploratory phase.
3. Objectives
The overall objective of this RTS is to ensure consistent application of measures taken by the person advising on or selling a PRIIP (hereafter the PRIIP distributor, who may also be the PRIIP manufacturer) to ensure that each retail investor receives the relevant KID sufficiently early – in view of his needs and the nature of the PRIIP – for the KID to be useful for the investment decision making of that retail investor. Consistency in approach taken by different PRIIP distributors would ensure retail investors can always trust in receiving a KID at the right time, and that these distributors are clear as to their obligations in respect of the PRIIPs Regulation. The policy issues to address therefore are:
 Degree of harmonisation of approach
 Adaptation to needs of the retail investor
 Adaptation to PRIIP
4. Policy options

134
This section explains the rationale behind the most relevant alternative solutions that the Joint Committee has examined when designing the RTS proposals. Policy issue 1:Degree of harmonisation of approach Option 1.1: Highest degree of flexibility for distributor Option 1.2: Common principles (minimum standards) Option 1.3: Detailed rules In order for the PRIIPs distributors to be in a position to be able to know when they have satisfied the obligation to provide the KID ‘in good time’, some questions arise as to the broad regulatory approach that might be adopted. At one extreme, it could be left entirely for each distributor to assess, according to criteria and processes of their own, what might count as ‘in good time’. At the other extreme, strict prescriptive rules might be established, setting out, for instance, the specific period of time (for instance hours or minutes, or a prescribed cooling off period of a day or two) during which the retail investor can reflect about the PRIIP. Another option would be to set in place some common principles and criteria, but to leave open. This is the approach already established in MiFID II as set out in Recital 83. The reference to MiFID II is important, in that establishing different standards in the context of the PRIIPs Regulation to those applying on distributors in respect of MiFID II could be a material cost driver for distributors, who would need to comply with different standards for different products (that is, those that are PRIIPs and those that are not PRIIPs). Policy Option 1.1: Highest degree of flexibility for distributors No. Pros Cons 1 Distributors able to achieve maximum proportionality to their customer base and its needs. Approaches of different PRIIPs distributors diverge, undermining retail investor trust. 2 - Some legal uncertainty/legal risk for distributors in assessing compliance. Policy Option 1.2: Common principles (minimum standards) No. Pros Cons 1 Greater legal certainty for distributors than option 1.1. Minimal standards may still leave room for divergence in approaches, with potential for some retail investors to receive KID too late in the process.

135
No. Pros Cons
2
Depending on substance of standards, permits greatest coherence with other distribution rules (MiFID II). Minimal standards may still leave room for some legal uncertainty/legal risk for distributors in assessing compliance. 3 Flexibility for distributors to adapt procedures -- proportionality in costs. Policy Option 1.3: Detailed rules No. Pros Cons 1 Greatest legal certainty for distributors. Lack of flexibility for retail investors if a long period is set for review of KID, yet retail investors very familiar with investment. 2 Lack of flexibility and disproportionate costs in some cases for distributors to adapt procedures for their products and typical customers. 3 Possibility of divergence between different rules (MiFID II, PRIIPs Regulation) – increasing costs. Policy issue 2: Adaptation to needs of the retail investor Option 2.1: Responsibility of the retail investor Option 2.1: Needs of generic retail investor Option 2.2: Needs assessed relative to the specific retail investor The amount of time a retail investor might need for reading and comprehending the KID is likely to vary significantly according to the capacities, experience and knowledge of that retail investor. Investors with considerable ongoing experience with specific instruments – for instance, those who have built up for themselves a portfolio of fund investments over many years – or past professional experience in such investments may require considerably less time to read and comprehend a KID. Other investors may need considerably longer. It can be argued that it would be up to the retail investors themselves to express their preferences. However, the obligation on timing applies to the PRIIP distributor, and it is up to the PRIIP distributor to assess what would count as ‘in good time’. When making such an assessment, a ‘one size fits all’ approach might be taken (setting the same standard for everyone, with sufficient time to suit those of lesser financial capability). On the other hand, a tailored approach, where the PRIIP distributor takes into account the retail investor and their needs, may be more proportionate.

136
Adaptation to needs is not only a reflection of the capacities, experience and knowledge of the retail investor, but also their expressed needs in respect of the specific transaction. A retail investor who wishes to buy a particular product at a particular price and time should not be unduly prevented from doing so, other regulation applying to the sale notwithstanding. Policy Option 2.1: Responsibility of the retail investor No. Pros Cons 1 Low cost and liability for distributors. Inconsistency with requirements under MiFID II. 2 Adaptability for some investors. Lower capability / experience retail investors may be less well informed. 3 - May be inconsistent with level one empowerment. Policy Option 2.2: Needs of generic retail investor No. Pros Cons 1 Simple standardized model potentially easy/cheap to implement for distributors. Potential inconsistency with requirements under MiFID II. 2 Legal certainty. Where standardized for lower capability retail investors, increases costs for higher capability or experienced retail investors. Policy Option 2.3: Needs assessed relative to the specific retail investor No. Pros Cons 1 May serve both lower and higher capability retail investors equally well. Some legal uncertainty for distributors. 2 Consistency with MiFID II May be more costly to implement. Policy issue 3: Adaptation to PRIIP

137
Option 3.1: Separate those PRIIPs with a comprehension alert and those without Option 3.2: Reflect specifics of different products Different PRIIPs may raise different challenges for retail investors. Some innovative or complex PRIIPs may be intrinsically more difficult for retail investors – irrespective of their experience, knowledge or capabilities – to understand and fully comprehend. While the KID itself is intended to provide as simple and easy to understand information as possible about a product, it cannot be assumed that all KIDs will be equally easy for retail investors to understand and compare. This implies therefore that for some PRIIPs the retail investor may need a longer time horizon to consider the information contained in the KID. Two basic approaches can be envisaged: a simplified approach, where the existence of a comprehension alert on the KID for a PRIIP triggers a mandatory requirement to allow a longer period for deliberations by retail investors, and a more flexible approach which leaves it for the PRIIP distributor to consider on a case by case basis the appropriate period needed. The latter is the approach proposed under MiFID II. In addition, it is not clear that all products that possess a comprehension alert or vice versa will be equally difficult or easy to comprehend for all retail investors, so removing all flexibility on this point may nonetheless reduce consumer protection levels. Policy Option 3.1: Separate those PRIIPs with a comprehension alert and those without No. Pros Cons 1 Simplicity, low costs of implementation. May be insufficiently tailored to specific retail investors and their needs. 2 Legal clarity. May be insufficiently tailored to the specific complexity of a specific PRIIP in the view of the distributor. 3 Inconsistency with MiFID II Policy Option 3.2: Reflect specifics of different products No. Pros Cons

138
No. Pros Cons
1 Adaptability for specific PRIIPs. Legal uncertainty, with some additional costs. 2 Retail investor needs more comprehensively taken into account 3 Consistency with MiFID II
5. Analysis of impacts
Rules on the timing of the delivery of the KID – derogations related to distance sales notwithstanding – are aimed at ensuring that distributors consistently take into account the varying needs of retail investors in relation to different transactions. By its nature, this entails a certain flexibility – putting the responsibility and liability for this onto the PRIIP distributor. A ‘one size fits all’ approach would be cheaper to implement and would of course carry less liability for the PRIIP distributor, but would be less likely to ensure retail investors get the protection that is sought for them by the level one Regulation. These considerations have already been taken on board by the co-legislators in regards the existing criteria established through MiFID II in Recital 83. Inconsistency with these criteria for the KID could create additional uncertainty and costs for PRIIPs distributors subject to MiFID II, while at the same time the solution established in MiFID II has also been designed in view of addressing the same policy issues as set out here in respect of the KID. For those subject to MiFID II, the baseline arguably – due to the need to comply with MiFID II already – contains the costs of applying these same standards in respect of the KID. For those not subject to MiFID II – those subject to the IMD – costs of implementation would include:
 establishing policies – where necessary in view of the size of the distributor – on timing of delivery and being prepared to explain the appropriateness of these;  establishing documentation, as may be requested by relevant supervisors;  gathering information on the needs, demands, knowledge and experience of relevant retail investors. The last point – in terms of ‘know your customer’ consequences – would however already apply under the IMD, such that the incremental costs of taking the customers’ needs, demands, knowledge and experience into account also with the timing of the delivery of information to the customer should be low.

139
Benefits
The effectiveness of the KID depends on it being provided or made available to retail investors prior to their investment decision making. For this reason, all measures that would increase the extent to which retail investors are able to read and use the KID, and the emphasis placed upon the KID by PRIIPs distributors, can be expected to reduce levels of mis-selling and mis-buying. The extent of this benefit however would depend on the overall effectiveness of the KID itself, and it is difficult to separate the benefit arising from timely delivery from the benefit arising from the quality and effectiveness of the document itself.
6. Monitoring indicators
The overall objective of this RTS is to ensure consistent application of measures taken by the person advising on or selling a PRIIP (hereafter the PRIIP distributor, who may also be the PRIIP manufacturer) to ensure that each retail investor receives the relevant KID sufficiently early – in view of his needs and the nature of the PRIIP – for the KID to be useful for the investment decision making of that retail investor. Consistency in approach taken by different PRIIP distributors would ensure retail investors can always trust in receiving a KID at the right time, and that these distributors are clear as to their obligations in respect of the PRIIPs Regulation. Monitoring indicators that have been identified are:
 Appropriate timing of delivery, as identified in mystery shopping exercises, including consistency in outcomes across different PRIIP distributors on delivery to similar types of customer  Appropriate use of concept of ‘urgency’ (including prior provision of KID in this case – no derogation from pre-contractual delivery)

140
4.2 Feedback on public consultation
4.2.1 First Discussion Paper

Background
A Discussion Paper (JC/DP/2014/02) was published for consultation on 17 November 2014, which closed on the 17 February 2015.. In total 87 stakeholders responded to one or more of the questions. The stakeholders included industry participants and associations from all sectors, including intermediaries, consumer representative organisations, the ESAs stakeholder groups, and public sector authorities and bodies.
Summary of responses
Preliminary feedback on those questions related to the summary risk indicator, performance scenarios and costs has already been included in the second (‘technical’) discussion paper (JC/DP/2015/01). Review, revision and republication of the KID
Chapter 7 of the Discussion Paper, and questions 49 to 52, related to the empowerment for the
ESAs to draft RTS under Article 10. In total 56 stakeholders responded to one or more of these specific questions. Almost all respondents agreed on building on the existing rules for the UCITS KIID, although most of them see the need for some adaptation to accommodate the broader range of products covered by the PRIIPs Regulation. Some manufacturers producing PRIIPs that might be placed on secondary markets were concerned that review, revision and republication requirements could be disproportionate, or raised the question as to whether the originating manufacturer of the PRIIP should be responsible. Statements from a wide range of types of stakeholders also highlighted challenges in specifying ‘materiality’ triggers for republication. Leaving these too open to interpretation could lead to different approaches, reducing comparability, while being too prescriptive could raise costs unnecessarily. None of the industry stakeholders was in favour of an active communication model regarding updates and existing retail investors. Consumer organizations viewed active communication as useful in some cases, for instance where there is a change to the cost structure or objectives of the PRIIP, or where its risk class changes.

141
It was noted that the manufacturer may well not be aware of who end investors are, for instance where units or shares of a PRIIP are held by distributors or other third parties, or are held through nominee accounts on systems where the end investor does not or cannot directly participate. The potential need for adaptation of the UCITS requirements has been taken into account by the ESAs when preparing the draft RTS. The ESAs note the difficulties with requiring an active communication model in respect of existing investors, also in view of the primary purpose of the KID as a pre-contractual rather than postcontractual communication document. Current developments of web-based publication and investment aggregation tools may, in the view of the ESAs, offer practical solutions for efficiently informing retail investors, and for keeping those retail investors up to date in regards to their existing investments. Timing of delivery
Chapter 8 of the Discussion Paper, containing questions 53 and 54, related to the empowerment
for the ESAs to draft RTS under Article 13. In total 54 stakeholders responded to at least one of these specific questions. A majority of respondents of all types agreed on building on the standard set in MiFID II Recital 83 as a model for the technical standards. Some mentioned that the standards should take into account that the complexity of a PRIIP might not in itself determine for all retail investors a necessity for additional time to understand that PRIIP – that is to say, the time needed should take into account the novelty for the retail investor. A number of respondents mentioned the importance of coherence with requirements under the Distance Marketing in Financial Services Directive, for instance in relation to ‘cooling off’ periods. Some mentioned the importance in their view of not impeding timely transactions where this is at the initiative of the retail investor. Recital 83 to the MiFID II sets out criteria to be taken into account, but allows flexibility over the combination or weighting of these criteria. The criteria include both the familiarity of the specific investor with the product or products of a similar kind, and the complexity of the product. Under this approach, it would be up to the person advising on or selling the PRIIP to consider, for instance, whether the complexity of a PRIIP outweighs the familiarity of the retail investor with investments of that kind. The requirements in the Distance Marketing in Financial Services Directive related to cooling off periods are not relevant in the context of the timing of delivery of pre-contractual information as set out in this empowerment, as these requirements relate to the time following the conclusion of a contract through distance means, where a KID will not have been provided ‘in good time before’ the retail investor is bound by the contract. This situation is covered separately in Article 13 (3),

142 which provides for a derogation to the requirement on provision ‘in good time before’. The empowerment that is the subject of this Consultation Paper does not relate to conditions in relation to the derogation set out in Article 13 (3). However, the text has been clarified in relation to cooling-off periods for the avoidance of doubt.
4.2.2 Second (‘Technical’) Discussion Paper

Background
A Discussion Paper (JC/DP/2015/01) was published for consultation on 23 June 2015, which closed on the 17 August 2015. In total 67 stakeholders responded to one or more of the questions. The stakeholders included industry participants and associations from all sectors, including intermediaries, consumer representative organisations, the ESAs stakeholder groups, and public sector authorities and bodies.
Summary of responses on Costs
Cost disclosure issues in relation to investment funds (list of costs) On the list of costs, most respondents addressed the question whether the list should be exhaustive or not. Diverging responses were given. The need to define further some types of costs was underlined by some respondents (acquisition costs, marketing and constitution costs, financing costs, including interest on borrowing). Some stakeholders were of the view that performance fees should be included in the summary cost indicator while others, including the majority of players from the fund industry, objected to this, and preferred to show performance fees separately. Some stakeholders advocate in favour of including carried interests in the ongoing costs figure, while others wish to present them separately. Some stakeholders are of the view that carried interests are not costs. Other stakeholders suggested including other types of costs in the non-exhaustive list of costs to be disclosed, such as listing fees. On transaction costs related issues, the majority of responses favoured the hybrid approach (between an actual costs based methodology and a standardized model). There was a lot of support for a standardised approach but more in the context of all of industry using the same methodology and approach to ensure consistency than having a set figure to use. Two responses stated explicit costs only should be taken into account. One response said that transaction costs should only be reported where they have a negative impact on the investment. Some

143 respondents from the insurance industry indicated that implicit transaction costs are marginal and negligible for the investment-based insurance market. Two respondents were of the view that the details of the calculation of transaction costs is too technical for a regulatory technical standard and should be left to level 3 guidelines in the event that changes are needed to the methodology once implemented. One response added that the standardised approach should be established by ESMA based on market data to be collected under the new MiFIR transaction reporting regime. In the view of this respondent, it should be updated in a regular way, probably annually, in order to fit into the regular revision cycle of the KID. One respondent indicated that transaction costs for property are very different from other securities – there could be costs in some years and none in others which makes them difficult to estimate in advance and information provided for one given year may not reflect the overall transaction costs over the time frame of the investment of a retail investor. Cost disclosure issues in relation to structured products (list of costs) Several stakeholders recognized that all costs of structured products are known by the manufacturer because these costs are all embedded in the price. Some respondents underlined that it is important to make a distinction between the costs borne by the manufacturer (e.g. hedging costs, legal costs, capital requirement costs) that are not passed to the client and the costs directly borne by the investor (e.g. sales commission). Several stakeholders stressed the need of having common and shared rules on the lists of costs in order to avoid arbitrariness and improve transparency and therefore that all costs should be described in details in the Regulatory Technical Standards. The respondents to the Technical Discussion Paper did not identify relevant differences between the costs of structured deposits and the costs of structured products. On the estimate of the fair value, almost all respondents agreed on the suggested principle-based approach, while all respondents objected to a very prescriptive approach. Cost disclosure issues in relation to insurance-based investment products (list of costs) On biometric risk premium (BRP) related issues, the majority of respondents from the insurance industry supported not including the full biometric risk premium in the summary cost indicator. Instead of giving information on biometric risk premium in the cost section of the KID, some stakeholders indicated they would prefer providing the related information only in the “What is this product?” section. These respondents considered that the BRP is a price and not a cost, and that including the full BRP in a cost summary indicator would not ensure a level playing field, having also in mind that in their view, BRP depends on the age of the investor and therefore cannot be disclosed accurately in a pre-contractual
document.

144
Some respondents outside the insurance sector favoured showing the full biometric risk premium as cost, considering that the costs included in the summary cost indicator shall comprise all the deductions from the investment. Overall the respondents did not elaborate further on the idea to split the BRP in a fair value and a cost part (part of BRP that exceeds the fair value). Regarding the list of costs of insurance-based investment products, several respondents from the insurance industry stressed that for insurance companies costs at the level of the company do not automatically generate costs at the level of the contract. On optional costs depending on the behaviour of the investor, several respondents stressed that these costs should not be considered, because these costs can be avoided by the investor and can hardly be anticipated by the manufacturer. On profit sharing, the feedback was quite inhomogeneous. Several respondents welcomed the fact that performance decreasing effects of profit participation shall be considered as costs. Some argued that a significant part of the return of assets are retained by insurers through profit participation mechanisms and therefore should be considered as cost. Others expressed the view that profits that are not distributed cannot be seen as costs and should be seen as a premium for the insurer for taking on risks. In the same vein, views on cost decreasing effects were split. Some stakeholders indicated that life insurers should not be allowed to consider parts of the profit sharing as deduction from costs, because in their views, future profit sharing is not guaranteed. Other respondents expressed the view that it is very difficult or even impossible to estimate the effect of future profit sharing on costs. One reason mentioned is that in several Member States there are profit sharing mechanisms at the level of the company that do not automatically translate into a certain profit sharing at the level of the contract. The effects on the single contract are therefore difficult to estimate. Cost disclosure issues in relation to the choice of the summary cost indicator A majority of the responses to the TDP expressed support for the Reduction in Yield (RIY) approach, as compared to the Total cost ratio (TCR) approach. The stakeholders that supported this approach were representing all types of sectors (fund and insurance industries, and to a less extent, the banking industry). Some stakeholders from the banking industry objected that the RIY approach implemented to structured products would raise technical issues. In that respect, some respondents were of the view that the assumptions on the future performances / returns should be consistent with the ones used in the performance scenarios section of the KID.
3. Summary of the feedback on Risk and Reward
I. Distribution of returns

145
On the general issue regarding the distribution of returns, most respondents do not actually make a distinction concerning how a distribution of returns should be established for the risk indicator or the performance scenarios. Some respondents point out that the approach followed for both indicators does not necessarily have to be identical, given the different intentions of both. Most respondents, especially those representing the investment fund or banking industries, prefer that estimates of return distributions should be based upon the distribution of returns directly obtained from historical data. Mostly respondents from the insurance industry would favour the application of stochastic modelling based on parameters estimated from historical data. Concerning the time value of money and the potential use of benchmarks, most respondents believe that performance should be measured against the amount invested without any adjustment of the real term value of the investment in the future. Some respondents suggest that, in particular for longer term products, a disclaimer could be included, pointing out to the effect of inflation. For those respondents who favour the use of probabilistic performance scenarios, a majority prefers applying the assumption that the assets grow at the risk free rate adjusted by an asset specific risk premium for the calculation of the performance scenarios. A minority of respondents view that risk premiums should not be used, neither for the purpose of performance scenarios, nor for the purposes of the risk indicator, given a potential bias of such estimates. Concerning the time frame that the Summary Risk Indicator and Performance Scenarios should be based on, most respondents prefer showing the risk indicator for the recommended holding period, but to include a warning or narrative text that explains the possible variation in risk over time. Still, a few respondents would favour showing the risk indicator and performance scenarios also for several intermediate times as well as the recommended holding period.
II. Risk indicator
Regarding the discussion of the risk indicators, the Technical Discussion Paper was focused on a number of specific issues. In particular, on the issue of the use of external credit ratings for the assessment of credit risk, most of the respondents are in favour of a qualitative approach (i.e. the use of credit ratings). A minority, mainly representing the structured products industry, would prefer a quantitative approach. The general preference may be due to the non-availability or volatility of quantitative data such as credit or CDS spreads. Proponents of the use of credit ratings underscore the importance of comparability and consistency of credit ratings with reference to the CRR and Solvency II since it would ensure a proper assignment of the various ratings to different credit quality steps. For cases where no external credit ratings are available, views are less clear. Some favour the application of the credit rating of comparable or peer companies, while others would also take into account whether the undertaking is prudentially supervised. Representatives from the insurance industry noted that prudential requirements under Solvency II, as well as the existence of insurance guarantee schemes (or deposit insurance), should be considered.

146
Regarding the assessment of liquidity in the risk indicator, most respondents view that liquidity matters mostly for trading clients and less for retail investors that follow a buy-and-hold strategy, as such it depends on the individual retail investor’s time horizon, preferences and circumstances. Thus, respondents would prefer a narrative or warning below the summary risk indicator. Some respondents note that the liquidity profile should be described under the sections “What is this product?” and “How long should I hold it and can I take money out early?”. Furthermore, concerning the qualitatively based risk indicator combining credit and market risk, most respondents do not prefer such an approach. In particular, many respondents are concerned that a quantitative rationale for the risk assessment would be missing. In addition, it may be arbitrary or not sufficiently able to differentiate between particular products. Few respondents note, however, that such an approach could be easier to implement and to monitor. The proposed indicator based on a quantitative volatility measure for market risk and a qualitative measure for credit risk is favoured by some respondents, mostly representing UCITS. Such respondents feel that it would be the closest approach to the current Synthetic Risk and Reward Indicator and could be easier to implement by manufacturers. Other respondents note that this option may not be suitable for long-term products, especially with a view to the insurance industry. One respondent proposes the combination of options 2 and 3. More than half of the respondents is not in favour of option 3, an indicator on quantitative market and credit risk measures calculated using forward looking simulation models. In particular, many respondents feel that the methodology of approach 3 is (unnecessarily) complex for both manufacturers and retail investors. Complexity of calculations may lead to the lower usability. In addition, option 3 may imply high implementation costs both for manufacturers and for regulators to supervise. More than one third of the respondents favour option 3. Such respondents feel that implementation effort is manageable due to the fact that it is already in place and most of the PRIIP manufacturers are calculating risk figures like VaR for their products. The responses to the proposed amendments to option 3 were mixed, a majority of respondents however do not favour the amendments suggested. Such respondents fear the amendments would lead to higher implementation costs and would need increased supervision. There is no consensus amongst respondents regarding the level of prescriptiveness of parameters and model. Most of the respondents do not support option 4, the two-level indicator, either noting its complexity or lack of clarity, as it is described in the Technical Discussion Paper. In addition, the majority of respondents believe that the definition of the scale of the risk indicator should remain consistent with
UCITS approach, which uses a scale of 1–7. Some respondents view that a continuous numeric scale resulting from converting each feature into points and then summing those points would be appropriate. In general, respondents urge the need that the indicator gives a fair comparison of the different PRIIPs, is easily understood by retail investors and efficiently implemented by PRIIP manufacturers.
III. Performance

147
Concerning performance scenarios, the Technical Discussion Paper discusses three possible approaches. Most respondents favour a deterministic, “what if” approach, over a probabilistic approach. Further, most would prefer a “what if” manufacturer choice approach. Respondents argue that illustrative scenarios are simpler and easier to understand for retail investors. Respondents also reject the use of combinations between “what if” and probabilistic performance scenarios. Regarding the issue how prescribed “what if” scenarios should be standardised, some respondents support the use of historic scenarios, especially for unit-linked products. However, others are firmly against the use of historical scenarios, as these may not be appropriate for other products. Predefined scenarios may limit manufacturers’ discretion and enhance comparability. However, at the same time, they may be difficult and too complex to standardise across the scope of all PRIIPs. Only a minority of respondents favours the consistent use of probabilistic scenarios. Regarding the possible determination of percentiles under a probabilistic approach, many respondents support the proposed 10/50/90 approach, to the extent that they also support such an approach at all. Some of the proponents of a “what if” approach allowing for manufacturer choices would support further Guidelines by the Joint Committee. Others would prefer a more prescribed “what if” approach already in the draft RTS, especially those representing the insurance industry. However, it became apparent from the consultation that different types of PRIIPs might warrant different treatment in the performance scenarios. Almost all respondents are against the inclusion of credit events in the performance scenarios, irrespective of the approach. Only a few respondents representing consumer organisations would favour their inclusion. It should be highlighted that the inclusion of credit events would only be appropriate for products such as credit linked notes where it is considered that credit events are a core characteristic of such products. Respondents from the insurance industry referred to the importance of insurance guarantee schemes, as well as prudential regulation under Solvency II in this regard. Divided views are given by respondents as to whether redemption events should be considered. A number of respondents, especially from the structured products industry, would only favour the inclusion of automatic early redemptions, while voluntary early redemption should be explained in the KID section "How long should I hold it and can I take money out early?”. Also regarding the question whether the performance in case of an early exit should be shown, views are divided. Respondents from the insurance industry as well as from some banks note that the consequences of an exit before the recommended holding period should also be indicated in the KID section "How long should I hold it and can I take money out early?”. Some
respondents view that the fair value is an adequate measure relative to future performances but may be difficult to compute for some scenarios. For future periods, the starting point of the scenario is uncertain and the retail investor may misinterpret the conditional expectation of the shown intermediate performance.

148
4.2.3 Final Consultation Paper
Background
The Consultation Paper (JC/CP/2015/073) containing the draft RTS was published on 11 November 2015 with a period to react for stakeholders ending 29 January 2016. In total 103 stakeholders and stakeholder groups responded to the questions The stakeholders included industry participants and associations from all sectors, including intermediaries, consumer representative organizations, the ESAs stakeholder groups, public sector authorities and bodies. General statements In reacting to the Consultation Paper most stakeholders take the opportunity to flag overall concerns with the RTS work and the process in the next months. The majority raise problems foreseen with the time to practically implement these RTS. They state that there will be not enough time left to produce KIDs for all PRIIPs after getting legal certainty around the detailed rules (when the RTS will be published in the Official Journal). Most manufacturers claim that they will need at least 9 months to one year to program the methodology in their systems, produce KIDs and test their compliance. An additional complication raised here is the absence of a grandfathering provision: there is no exception for PRIIPs produced in the past, not actively marketed anymore, but still available on secondary markets. For some manufacturers producing KIDs for these PRIIPs is a considerable additional task. 12 22 21 17 15 5 11 consumer representatives insurance and pension asset management banking service providers regulated markets other

149
The ESAs acknowledge most of the practical problems raised, but are not in the position to change the timeline or scope of the PRIIPs Regulation. They did and will flag these practical issues to the European legislators and will try to accommodate a smooth implementation by helping with Q&As and other guidance in offering solutions for practical problems. Some stakeholders pointed out challenges with applying the KID regime to Exchange Traded Derivatives, especially due to their fast changing risk and performance characteristics. The ESAs decided to allow for a more generic presentation for these products, explaining the basic principles of their expected behavior under different market situations. Comprehension alert Question 1 Would you see merit in the ESAs clarifying further the criteria set out in Recital 18 mentioned above by way of guidelines? Most respondents would see merit in the ESAs clarifying the criteria for the use of a comprehension alert, but also acknowledge that there is no specific mandate to do so in the RTS. A common approach would be welcome though, as the current criteria in the PRIIPs regulation are regarded as not very helpful. Guidelines are seen as the most obvious tool for this. Standardised amounts used as a basis for calculations Question 2 (i) Would you agree with the assumptions used for the proposed default amounts? Are you of the opinion that these prescribed amounts should be amended? If yes, how and why? (ii) Would you favour an approach in which the prescribed standardised amount is the default option, unless the PRIIP has a known required investment amount and price which can be used instead? Feedback on the proposed default amounts showed a strong support for setting a same amount as the default amount for all PRIIPs products (except in case of regular premiums). This approach was favored by a very large number of stakeholders. These respondents indicated that this solution would allow for better comparability between the different types of PRIIPs. Beyond that, a majority of stakeholders suggested setting the default amount at € 10.000. These respondents were of the view that this amount would give a clear picture of the typical amount invested in a PRIIP. As a consequence, the ESAs decided to define a single default amount of € 10.000 and decided to set the regular premium default amount at € 1.000. The ESAs took note of the comments of some stakeholders that the default amount should be set at national level. The ESAs underline that this approach would not be in line with the single market concept. The ESAs also took note of a few requests from some respondents asking for a more flexible approach, e.g. by fixing different default amounts per product category. The ESAs

150 are however are of the view that a more flexible approach would not be in line with the objective of comparability between the different types of PRIIPs. Risk and Reward section Application of the methodology Question 3 For PRIIPs that fall into category II and for which the Cornish Fisher expansion is used as a methodology to compute the VaR equivalent Volatility do you think a bootstrapping approach should be used instead? Please explain the reasons for your opinion? Question 4 Would you favour a different confidence interval to compute the VaR? If so, please explain which confidence interval you would use and state your reasons why. Question 9 Are you of the opinion that for PRIIPs that offer a capital protection during their whole lifespan and can be redeemed against their initial investment at any time over the life of the PRIIP a qualitatively assessment and automatic allocation to MRM class 1 should be permitted? Are you of the opinion that the criteria of the 5 year tenor is relevant, irrespective of the redemption characteristics? Question 13 Are you of the opinion that the current Consultation Paper sufficiently addresses this issue? Do you it is made sufficiently clear that the value of a PRIIP could be significantly less compared to the guaranteed value during the life of the PRIIP? Several alternatives are analyzed in the Impact Assessment under policy option 5: do you see any additional analysis for these assessment? Concerning the methodology, a distinction is proposed for some categories of products, to which should apply a Cornish Fisher methodology, and for some other categories of products, to which should apply a bootstrapping approach. For PRIIPs that fall into category II and for which the Cornish Fisher expansion is used as a methodology to compute the VaR equivalent Volatility, stakeholders were asked whether a bootstrapping approach should be used instead. A significant majority of stakeholders believe that the Cornish Fisher expansion is an appropriate methodology. Most of them felt that while the two methodologies provide similar results, the bootstrapping approach would be disproportionally complex and imply higher implementation and on-going costs, and could therefor agree to a different methodology for different categories of PRIIPs. Few stakeholders found that the bootstrapping approach should be favoured, primarily for consistency reasons with the methodology used for other categories. As a consequence the ESAs decided to maintain the proposal for using a bootstrap methodology and the Cornish Fisher methodology. Further they ESAs have revised the text to address more clearly what products would be considered which categories PRIIPs. With respect to the question of the confidence level a majority respondents stated that they did not agree with the proposed confidence level. About a third of the respondents did agree with the proposed confidence level. From the respondents that did not agree to the confidence
level interval a majority stated that they did not agree with the proposed methodology in general. The suggestions on what the confidence interval should be were mixed. Most respondents who

151 disagreed with the confidence interval proposed to lower the confidence level (e.g. 1% or 0,5%). Some respondents appreciated a higher confidence level (e.g. 5% or 10%) and some suggested a flexible approach to allow to set the confidence level per country, industry or type of PRIIPs. As a consequence the ESAs decided to maintain the confidence level as proposed. The ESAs underline that more flexible approaches would not be in line with the single market concept. Although there was a majority in favor of a different confidence interval, the respondents diverged significantly in what this interval then should be. No majority for a different interval could be found. The MRM in the SRI show the risk of the product at the recommended holding period, assuming the investor holds it until maturity. A vast majority of the respondents agreed that this should be clarified in the KID to retail investors Most of these respondents were in favour of explaining the risk in a warning or narrative just below the SRI. A minority prefers to disclose such information in another section of the KID. A minority of the respondents did not agree with the principle of calculating the risk at the recommended holding period. Of these respondents not in favour some suggested to show multiple time frames in the SRI or to prescribe the recommended holding period. The methodology for the MRM, as proposed in the CP, allows a qualitative assessment for PRIIPs with capital protection as long as their tenor is up to 5 years. A large majority of respondents were not in favour of establishing a 5 year limit to distinguish products, considering it as arbitrary, not realistic or hampering comparability. Respondents showed mixed views about the general approach to allow a qualitative criteria to assign market risk class 1 to products with 100% guarantee (irrespective of the tenor). An important number of respondents (particularly in the insurance sector), defended this approach considering that inflation should not be considered in the market risk measure. However, there were also a number of respondents that opposed to that qualitative criteria, considering that tenor mattered, in order to take into account inflation and avoid monetary illusion or to consider developments in the market and impact in case of early exit. Some voices opposed the end of maturity and recommended applying a uniform holding period for all products. Given the responses that the tenor was considered to be arbitrarily selected and that the number of PRIIPs that could make use of the qualitative criteria limited by a tenor would not be substantial, the ESAs came to the conclusion that the tenor should not be applied. The ESAs appreciate the difficulties stated by a majority of the respondents. Especially with regards to the decrease in value of money over time is considered as a risk that should be taken into account. This aspect is taken into consideration when quantitatively calculating for the market risk
classification as well. Providing that an equal treatment is crucial the qualitative assignment to risk class 1 for capital protected PRIIPs was deleted from the RTS. The introduction of the qualitative assignment was mainly in order not to create unnecessary burden for PRIIPs that offer an unconditional protection of capital. Additional provisions have been made to accommodate these types of PRIIPs by allowing a simplified calculation in the current RTS. Summary Risk Indicator Question 6 Would you favour PRIIP manufacturers having the option to voluntarily increase the disclosed SRI? In which circumstances? Would such an approach entail unintended consequences? Question 7 Do you agree with an adjustment of the credit risk for the tenor, and how would you propose to make such an adjustment?

152
Question 8
Do you agree with the scales of the classes MRM, CRM and SRI? If not, please specify your alternative proposal and include your reasoning. Regarding the possibility to allow manufacturers to voluntarily increase the risk indicator, respondents showed mixed views. Around half of the respondents considered that this could be a useful approach if the product has special features that are not considered in the methodology, though some of them recommended to limit the option to certain circumstances. The other half of the respondents did not support this option, being the more repeated argument that it would reduce comparability as it may imply different risk indicators for similar products. The ESAs agreed with the respondents opposing the possibility to voluntarily increase the risk indicator, that this might lead to less comparability. Further we agree with the concern some respondents made that this might increase uncertainty for market participants and that it would be difficult to foresee all possible unintended consequences of this option. Therefor no possibility is given to voluntarily increase the Risk indicator. With respect to the question whether the credit risk measure should be related to the tenor of a PRIIP the respondents’ positions are clearly divided. A minority of the respondents did not agree with the principle of integrating the MRM and the CRM and therefore did not comment on this specific question. Other respondents stated that the CRM is not applicable to them. For the remaining respondents about half of them did and half of them did not want to amend the CRM for the tenor of the PRIIPs. An argument in favor of amending for tenor were that the longer the term of the PRIIP the more relevant the credit risk of the PRIIP becomes. An argument against the amendment is that the application of the methodology (qualitative) does take such tenor already into account. Some respondents suggested to apply the default probabilities related to a credit rating as a way to measure credit risk. There were different suggestions from the ones in favor of amending the CRM for tenor (e.g. up to 5 and more than 5 years or 0-1;1-5;>5 years). Some respondents mentioned that amending the tenor should be done by at least lowering the impact of the CRM compared to the MRM in total weighting of the SRI for relatively smaller tenors. The ESAs have carefully considered the reactions of respondents with regards of taking into account the tenor of the PRIIPs and adjusting the tenor. Some respondents pointed out that ECAI also take into account the aspect of time when applying a credit rating. This was also flagged under the comment made on the Credit risk measure in general. The ESAs then considered that the credit quality steps could be aligned with the default probabilities and so taking into account the aspect of time. With respect to the question on the different scales many responses have been received, we address them separately. The remarks on Credit
Risk are addressed in the following section. Market Risk Measure Regarding the market risk measure, the Consultation Paper suggests assigning market risk classes according to a volatility equivalent table that is largely in line with the calibration of the corresponding table for the Synthetic Risk and Reward Indicator in the UCITS KIID. About a third of the respondents have commented on the Market Risk scale.

153
Generally, while some respondents, often representing UCITS, support the suggested approach to the calibration of the MRM classes, there are many respondents, from other market segments as well as consumer representatives, who would prefer a different calibration. Among those that agree with the general calibration, few respondents would prefer reserving MRM class 7 for products where the retail investor may lose more than his initial investment. Those that prefer a different calibration view that the suggested calibration would be too conservative and not allow a sufficient differentiation between some products. In particular, the calibration would lead to too many products classified in MRM classes 6 or 7. Alternative calibrations have been suggested by this group of respondents. Representatives of the insurance industry voiced general concern on the overall approach to the MRM. The ESAs have carefully considered the suggestions on the recalibration of the MRM. In view of the responses received, the ESAs adjusted the calibration of the MRM taking into account the remarks made by market respondents. Agreeing that the UCITS scale tended to overly discriminate between lower risk products and under discriminate The changes to the scale are believed to be necessary and enable discriminating between reasonably volatile equity and more speculative equity to meet the investment reality better. This holds that the risk classes are redefined compared to the proposed calibration under the UCITS scale, allowing for a better overall discrimination. This also entails that some products moved to a lower risk class for the MRM than was proposed in the consultation paper. Risk class 7 is not purely reserved for products where investors could lose more than their investment as was suggested by some market respondents. The ESAs agree that very risky products should be assigned the highest risk class, but from a consumer perspective highly volatile products where you could lose all of your investment are also believed to be very risky. Reserving risk class 7 only for derivatives and products with residual loss and not speculative types of products was not considered sufficient from a consumer perspective. It would also hold that highly volatile products would be classified into a lower risk class, leaving less risk classes to discriminate between more volatile products. About a third of the market respondents made remarks about the default classification into the MRM, in case there was no sufficient data to calculate according to the MRM methodology. Participants stated that it seemed arbitrary and especially not granular enough how the classification by lack of data was formulated in the RTS that was consulted. Together with that argumentation the position of these products compared to the classification based on the calculation was considered not fair, e.g. the risk profile of products with lack of data was considered to be misclassified. The ESAs have carefully
considered the concerns raised by the market respondents regarding the Category 5 products. A more expanded table could meet the concerns on the granularity of the respondents however this would require to further define asset classes. Further this assigning of risk classes is likely to be considered arbitrary as well not satisfying the other concerns from

154 market respondents. Therefor the ESAs made an effort to reduce the group of PRIIPs that would have to rely on an automatic classification. For being able to make a quantitative basement a certain amount of data should be present. The ESAs have also taken into account the numerous comments made about the lack of clarity about the data frequency that should be used and the circumstances where the use of benchmarks and proxies could be used. The intention to allow for weekly data was not clear from the draft RTS. The ESAs have adjusted the RTS to clarify this and also on what historical information can or must be used, and further description as to the use of benchmarks and proxies. It is clarified that the minimum amount of data points is set at 120 to ensure a certain statistical assurance. This holds that at least 2 years of historical daily prices, 4 years of historical weekly prices or 5 years of bimonthly proxies should be available in order to have a sufficiently accurate calculation according to bootstrapping (category III) or Cornish Fisher (Category II). The RTS also provides that when such data is not available, existing appropriate benchmarks or proxies shall be used to run the market risk assessment. This would reduce the number of PRIIPs relying on automatic (qualitative) classification already significantly. However there are still PRIIPs that will not have sufficient data. For PRIIPs having monthly data about 60 data points are available, this increase the chances of misclassifying PRIIPs. For this bigger chance of error, an increase of 1 risk class is required. Reducing the number of PRIIPs relying on category 5 even further, and providing more granularity in classifying the risks. Finally, the remaining PRIIPs which cannot meet the data requirements for the quantitative measures are automatically classified into MRM class 6. Aggregation of CRM and MRM into SRI In just under half of all respondents to the CP expressed their concerns on the aggregation of CRM and MRM into SRI. Stating their doubts on the understanding of retail investors of these very different risk. About a third of the respondents to this part of Question 8 stated that MRM and CRM should not be integrated, some of these parties state it should only be mentioned in a narrative and others are in favor of disclosing separate indicators. There were very differing views on the weighting of Credit Risk in the SRI. About half of the respondents responding on the SRI scale, mentioned the weighting of Credit risk in the aggregated SRI. A minority of those comments stated that the Credit risk was considered to be overweighed in the SRI and a majority of these respondents came to the conclusion that the credit risk is underweighted. The overweighting of CRM in the SRI was specifically mentioned in regards to the beginning of the scale, specifically because the first three Credit Quality Steps do not lead to an impact on the SRI. , The fact that CRM4 implies a jump of 2
classes in the overall SRI, and that CRM4 and CRM5 does not imply a difference, is also mentioned by several respondents. This is also related to the comments made on the CRM. On the other hand a significant number of respondents flagged that especially at the end of the scale Credit Risk is underweighted, masking the Credit Risk when the PRIIP has a higher MRM classification. Some propose to take instead the mean of both measures. We have carefully taken into consideration the comments made by the market respondents and reviewed them also in light of their overall raised concerns and comments. By doing so the issues

155 flagged on integrating the scales mainly concerned the CRM in itself as the ESAs draw the conclusion. Given the views on the CRM which were considered strongly divided between sectors, the ESAs consider no further changes to the aggregation of CRM and MRM are necessary, given the adaptations made to the CRM are intended to provide a balanced and compromise of the Credit risk measurement between different sectors as well. One aspect that is recurring or coming from all groups is the irregularity of the proposed scale and especially that no distinction is made between AAA, AA and A ratings, they all give CRM 1, which holds that BBB is 3, this is what often is mentioned with the underweighting of CR. Credit Risk A majority of the respondents is of the opinion that the existence of a guarantee scheme should not be taken into account when assessing the credit risk of a PRIIP. Several respondents argue that the KID has been designed to make retail investors aware of the potential risks and benefits of/within a particular PRIIP. However, most respondents are of the opinion that retail investors should be informed about the existence of a compensation or guarantee scheme, either in a narrative or in the section ‘What happens if [name of the PRIIP manufacturer] is unable to pay out?’. A (small) minority of the respondents believe the existence of a guarantee scheme should be considered a credit risk mitigating factor and should as a consequence impact the CRM. The ESAs agree with the argumentation that compensation and guarantee schemes are not an intrinsic part of the product, but rather a part of the overall environment in which the PRIIP is made available. Other stakeholders state that external compensation schemes may vary between Member states, may change over time and their existence and scope may be unrelated to the specific risk profile of the product. Therefor the ESAs have maintained the proposal as it was and therefore not included guarantee schemes as mitigating factors for credit risk. In the narrative of the risk section the presence of a guarantee scheme is pointed out the retail investor and refers to other sections in the KID. Slightly less than half of the respondents are not aware of any other circumstances in which the credit risk assessment should be assumed to be mitigated, other than an appropriate collateralization of the credit risk, based on segregation of the assets. Some respondents (representing mainly the Asset Management sector) state that they welcome the proposal that, in principle, credit risk shall not be assessed on AIFs and UCITS and a minority of them particularly outline that they would argue with the principle of the look through approach with regards to structured funds. About half of the respondents agree with the proposed RTS that certain credit risk mitigating and escalating factors should be taken into account, such as subordination of claims or collateralization. Several respondents (representing mainly
the insurance sector) are of the opinion that the existence of prudential regulation, mainly Solvency II, should be taken into account when assessing the credit risk of a PRIIP. Such respondents are of the opinion that, in case the manufacturer abides by the requirements as set by prudential regulation, credit risk should either not be integrated in the risk indicator at all or that these PRIIPs should be automatically classified in credit risk class 1. However, a number of respondents argued that the existence of prudential regulation should not justify any mitigation of their credit risk assessment. It is argued that Solvency II Directive, CRD

156
IV/CRR and the Bank Recovery and Resolution Directive (among others) are merely tools to mitigate systemic risk by large institutions on the financial sector to reduce the need of public interventions for these institutions in times of great financial distress. These directives merely provide for further systemic risk buffers, but do not make these institutions infallible, thus rendering null and void the need to assess credit risk for PRIIPs connected to these institutions. The ESAs have carefully deliberated upon the suggestions made by market parties and appreciated the possible additional mitigating factors that were proposed including mitigating factors under prudential regulation. Also for providing more granularity to the automatic classification in case no ECAI rating is available. Although the prudential regulations might provide significant information, in the current situation the harmonization of the prudential regulations cannot be guaranteed at this point. At this point the consequences of a capital ratio for the possible loss to the investor in case of a default cannot be assessed adequately. In the future when the PRIIPs RTS is reviewed this could be a promising factor that should be considered. The ESAs have not dismissed the suggestions under prudential regulation though but taken them into account in a less ‘quantitative’ manner. The ESAs believe a delicate balance is crucial given the diverging views of different sectors. The ESAs have changed the draft RTS accordingly to account for the concerns raised from market parties. The use of the ECAI ratings where available is maintained, more granularity is added to the situation where no ECAI rating is available, which takes into account the fact whether an entity is subject to prudential regulation. Further ‘credit risk mitigating’ factors have been introduced, following the Market responses on segregation of assets, value of the assets held, and the precedence of claims. The ESAs agree with market respondents that these factors mitigate the risk of loss for retail investors in case of a default. Also escalating factors are introduced where this loss for the retail investor is not considered to be limited by the set-up of the PRIIP. A vast majority of the respondents believe the look through approach to the assessment of credit risk for a PRIIP packaged into another PRIIP is appropriate where the credit risk of the underlying PRIIP is passed to the retail investor. One respondent suggests to use an alternative approach, namely the CRR double default risk methodology. The ESAs therefor maintained the look through approach in the RTS. Currency risk A very large majority of industry respondents is not in favour of incorporating currency risk in the synthetic risk indicator when a PRIIP is issued in a different currency than the currency of the target market where the PRIIP is marketed. This requirement is seen as an additional burden on manufacturers operating in a cross border
context that would increase implementation costs. Also, it is argued that such currency risk is not intrinsic to the PRIIP but rather an investor or target market specific element. Moreover, its integration would render the construction of the synthetic risk indicator even more complex. It is nevertheless acknowledged that this is an important piece of information that should be mentioned in the narrative under the synthetic risk indicator. Several respondents proposed a redraft of the narrative in order to avoid that a different narrative and a separate KID should be used for every country where the PRIIP is marketed. About half of consumer representatives are also not in favour of the integration of this type of currency risk, as it is an element not intrinsic to the PRIIP. A few industry respondents and approximately the other half of consumer representatives are of the opinion that the synthetic risk indicator should take into account this type of currency risk to ensure investor protection, because exchange rate fluctuations may lead to substantial losses and because it is hard to evaluate for retail investors. Given the responses the ESAs have concluded

157 that the proposal on currency risk should be maintained. Holding that the currency risk will be addressed in the Risk and reward section as a narrative. Performance Generally, a number of respondents ask for more clarity regarding the calculation as well as the presentation of performance scenarios, for example, regarding roles and responsibilities of manufacturers, definition of holding periods etc. However, it is also noted that there is only limited time for the implementation of rules relating to PRIIPs and that guidance should not come too late. The Consultation Paper suggests that performance scenarios should be shown net of all costs and take into account performance fees. Especially respondents from the funds industry make the general comment that past performance should be included in the PRIIPs KIDs. With regard to whether performance fees should be used as a basis for calculations in the performance section13, respondents’ views are mixed. While especially representatives of the funds industry point out that performance fees may be taken into account, representatives of other industries view that this would not be applicable to them. Others opine that performance fees should not be taken into account in any case. Some respondents also raise the issue that performance fees may not only be paid in the moderate or favourable scenarios but also in others, as long as the product outperforms its benchmark even in a difficult market environment. Regarding whether the same benchmark return should be used for calculating performance fees for the unfavourable and favourable scenarios, or another approach should be used, respondents’ views are mixed as well. While some support is present to use the same benchmarks, others prefer the use of different benchmarks for the unfavourable and favourable scenarios and would also leave some discretion to manufacturers. A number of respondents would prefer if the recognition of performance fees would only kick in in the favourable scenario. Concerns are raised that by using data for the past five years that this would assume that the performance in that period may not be a suitable default assumption for the calculation of forward-looking performance scenarios. The ESAs recognized that when performance fees are triggered these should be taken into account. However the performance scenarios selected should not be driven by the triggering of the performance fees or not. Further changes have been made on the performance scenarios, making this requirement redundant. Concerning the presentation of the performance scenarios, the Consultation Paper proposes only to use tables. Regarding whether the presentation of the performance scenarios in the form of a graph should be preferred, or both a table and a graph, respondents’ views are split between only using a table or a graph. However, there is only little support for including both a table and a graph, mainly due to the possibility of causing confusion as
well as exceeding the available space in the KID.

13 I.e. calculate the return of the benchmark for the moderate scenario in such a way that the return generates the performance fee as prescribed in the cost section.

158
Regarding the use of graphs, a particular concern that is raised by some, is that the inclusion of interim periods may be misleading and give a wrong impression of the possible volatility of products. Given the diverging views between presenting tables or graphs and taking into account the issue of intermediate periods to be shown in a graph the ESAs have decided to provide a table in the KID. Several respondents also opine that for products with fixed maturities (incl. insurance products) no intermediate periods should be shown. Moreover, some respondents from the insurance industry are skeptical regarding the inclusion of biometric features and also prefer an even number of performance scenarios. Profit sharing should be included in the performance calculations, where relevant. In terms of general comments, more guidance and prescriptiveness are favoured by a number of respondents. A few respondents propose the use of percentiles in order to make performance scenarios more prescriptive. One respondent advocates using an alternative approach for the performance presentation of derivatives. Further, the timeframe of interim calculations should allow for flexible time periods that are displayed and it should be possible to use longer interim timeframes especially for long-term products. On the methodology aspect the ESA’s have considered the comments seriously and have deliberated on several alternatives in trying to achieve a more prescriptive method for the performance scenarios based on the following starting points;

Making sure the calculation was linked to the SRI calculation to prevent adding
burdensome and costly calculations to the SRI.
Making sure the calculation would result in a fair representation of the assumed risk and
reward trade-off.
Also no assumptions should be required on the risk premiums for different assets. This
would require prescriptive assumptions to be made on market developments and no straightforward or non-contentious methodology is available to do so. Therefore, risk premiums would be very hard to determine and are considered to result into an unstable methodology, for it would be subject (and more sensitive) to change over time. This eventually led to the proposal of using the Distribution of Returns (DoR) of the product generated for the MRM to select the performance scenarios from. The performance scenarios will be selected at the 10th, 50th and 90th percentile. The DoR following from Category III products will be based on the actual return of the historical data. These data will be measured at the end of the Recommended Holding Period. As there is no need for discounting there will be no correction for the risk free rate (as is the case for the MRM-calculation for category III PRIIPs). The ESAs feel that with this proposed methodology the concerns of market parties raised on this aspect are sufficiently addressed. Cost section

159
Transaction costs
Question 16
Do you agree with the scope of the assets mentioned in paragraph 25 of Annex VI on transaction costs for which this methodology is prescribed? If not, what alternative scope would you recommend? Question 17 Do you agree with the values of the figures included in this table? If not, which values would you suggest? (please note that this table could as well be included in guidelines, to allow for more flexibility in the revision of the figures) The transaction cost section received feedback on a number of areas. In general respondents agreed with the scope of assets for which the transaction cost calculation methodology is proposed. Asset management respondents tended to disagree with the methodology selected. They also raised a number of technical concerns about its feasibility. From the alternative viewpoint consumer organisations supported the approach and one stakeholder provided the view that the methodology was largely feasible. The ESAs considered the technical feedback and have revised the text to address the technical concerns raised, including by making certain terms more precise, and by making it clearer under which circumstances widely available prices such as opening and closing prices may be used. The text has also been revised so as not to require firms to use intra-day prices when calculating information for the period prior to the application of the regulation. Where the feedback disagreed with the policy position on transaction costs, the ESAs noted the feedback, but believe that there is no reason to revise the policy intent of the proposal, which is to identify the indirect costs associated with transactions in underlying investments. In terms of the standardised table, respondents (mainly asset managers or their representatives) expressed a positive view arguing that manufacturers should be able to use the standard table for products other than new PRIIPs. They were generally of the opinion that the numbers in the standardised table should be part of Level 3 guidelines rather than contained within the Regulatory Technical Standards. Some respondents did not believe all manufacturers should be compelled to use the standard table for 3 years. There was further feedback that the table is rather too simplistic and that there need to be more asset classes, and further discrimination for example between smaller and larger transactions, or to take account of more or less volatility in markets. The ESAs considered this feedback and have revised the criteria for using the standard table to allow firms to use some actual data after a PRIIP has been operating for 6 months. In terms of providing more granular data, the ESAs have added some additional asset classes, but consider that, since the table is to estimate the costs associated with new PRIIPs, it is not necessary to provide a more detailed breakdown, as there are other estimation factors as well as the estimated costs per transaction. The ESAs do not consider that this less
granular approach would be possible if there was more widespread use of the standard table, and for this reason do not agree with feedback proposing that the standard table is used for other purposes.

160
Biometric risk premiums
Question 19
Do you think that estimating the fair value of biometric risk premiums as stated in paragraph 55(b) of Annex VI would raise any technical or practical difficulties? Question 20 Knowing that the cost element of the biometric risk premium is included in the total costs calculation, how do you think the investor might be most efficiently informed about the other part of the biometric risk premium (i.e. the fair value), and/or the size of biometric risk premium overall? Do you consider it useful to include the fair value in a separate line in the first table, potentially below the RIY? Or should information on the fair value be disclosed in another part of the KID (for instance, the “What is this product?” section, where the draft RTS currently disclose biometric risk premiums in total, and/or in the performance section)? What accompanying narrative text do you think is needed, and where should this be placed, including specifically narrative text in the cost section? On biometric risk premium (BRP) related issues, the majority of respondents from the insurance industry still supported not including the full biometric risk premium in the summary cost indicator. Instead of giving information on biometric risk premium in the cost section of the KID, most of these stakeholders indicated they would prefer providing the related information only in the “What is this product?” section. These respondents considered that the BRP is a price and not a cost, and that including the full BRP in a cost summary indicator would not ensure a level playing field, having also in mind that in their view, BRP depends on the age of the investor and therefore cannot be disclosed accurately in a pre-contractual document. Many respondents outside the insurance sector favoured showing the full biometric risk premium as cost, considering that the costs included in the summary cost indicator shall comprise all the deductions from the investment. On the split of the BRP in a fair value and a cost component (part of BRP that exceeds the fair value), some respondents pointed out that the precise method to calculate the fair value of the BRP might need to be specified in level 3 guidelines. Overall, the respondents did not indicate that the estimate of the fair value of the BRP would raise significant technical difficulties. Having regard to this overall feedback, the ESAs decided to keep the approach suggested in the consultation paper on the estimate of the cost of BRP. The ESAs also clarified the presentation of the information on BRP in the cost section, including a dedicated line on the costs of the BRP in the second table on the breakdown of costs, and adding specific narratives on the the additional impact of the insurance premium payments. Monetary amounts Question 18 Do you agree that the monetary values indicated in the first table are a sum of costs over the respective holding periods? Or should the values reflect annualized amounts? If you prefer annualized amounts,
which method for annualisation should be used (e.g. arithmetic average or methods that consider discounting effects)?

161
Amounts to show in table 2 (breakdown of costs) Question 21 Given evidence as to the difficulties consumers may have using percentage figures, would you prefer an alternative presentation of the second table, solely using monetary values instead? As with the first table, please also explain what difficulties you think might arise from calculating monetary values, and whether this should be on an annualized basis, and if so, how? Alternative presentation of cost breakdown Question 22 Given the number of tables shown in the KID, do you think a more graphic presentation of the breakout table should be preferred? Question 23 The example presented above includes a possible way of showing the variability of performance fees, by showing the level for all three performance scenarios in the KID, highlighting the ‘moderate‘ scenario, which would be used for the calculation of the total costs. Do you believe that this additional information should be included in the KID? Question 24 To reduce the volume of information, should the first and the second table of Annex VII be combined in one table? Should this be supplemented with a breakdown of costs as suggested in the graphic above? Presentation format Question 26 Regarding the first table of the cost section presented in Annex VII, would you favour a detailed presentation of the different types of costs, as suggested in the Annex, including a split between one-off, recurring and incidental costs? Alternatively, would you favour a shorter presentation of costs showing only the total costs and the RIY? Question 27 Regarding the second table of the cost section presented in Annex VII, would you favour a presentation of the different types of costs showing RIY figures, as suggested in the Annex, or would you favour a presentation of costs under which each type of costs line would be expressed differently, and not as a RIY figure -expressed as a percentage of the initial invested amount, NAV, etc.? With respect to the question whether the sum of costs or annualized amounts should be indicated in the first table of the presentation format of the cost section, ESAs received diverging responses. Consumer representatives and some industry players argued in favour of disclosing the sums of costs. These stakeholders argued that the sum of costs would allow for a better overview of costs for investors, and would give a clear indication of costs over time. However, other industry players supported the idea of annualized values, in order to avoid giving the

162 impression that higher costs are incurred the longer a fund is held. They explained that, when using sums, the relative decline in one-off costs would not be sufficiently visible. The ESAs concluded that the sum of costs would be preferable to give a complete picture of costs over different holding periods and better reflect the L1 requirement of showing total costs. As to the question whether a more graphic presentation of the breakdown table (second table of the cost section) would be preferred, a clear majority of respondents favoured a table presentation because this provides clarity and allows for an easier comprehension of cost breakdown/composition. In line with the feedback, ESAs decided to keep the table presentation. In response to the question whether additional information should be given on performance fees (beyond the use of the value of the moderate scenario for the costs calculation) a strong majority of respondents expressed support for the approach presented in the CP. According to these stakeholders additional information would result in a situation where too much information could be given, potentially creating confusion and lowering the level of comprehensibility and comparability for the investor. In line with the feedback received, the ESAs did not change their proposed approach on performance fees in the cost section. With respect to the question whether the values in the tables should be in percentage or monetary terms, some consumer associations and some industry players advocated for a presentation of monetary values, while others advocated for a presentation in percentage terms. Other industry players supported an indication in percentage terms, referring to the current practice for UCITS and to the potential use of the KID in the context of cost calculations by the distributors. With respect to the second table, these stakeholders either supported a RIY figure or asked for including actual costs in percentage terms. The former recognized that the RIY figure allows for better consistency with the first table. The ESAs agree that consistency is important in order to enable the consumer to understand the presentation. The KID is a precontractual document for the investor which should help him/her to take an investment decision. Therefore the ESAs decided to show monetary and RIY percentage terms for the global costs in the first table, and provide for granularity of the RIY value at the recommended holding period in the second table. Many stakeholders also expressed support to the merging of the two tables in order to reduce the volume of information. However, a significant minority indicated they want to keep the two tables, because they are of the view that the overall information is valuable and keeping the same level of information in one table would be difficult. The ESAs agree that information provided in a combined table would be too complex. The ESAs took also into account feedback on the question whether the
first table should show a detailed presentation. Voices in favour and against these proposals were almost balanced. Those in favour insisted on the value of the information provided by a detailed presentation and oppose to hiding a part of information, whereas voices against feared that there is not enough space in a 3-pages document as the KID. Taking into account the feedback received on these two questions, the ESAs decided to keep the two tables. However, the first table is streamlined (just indicating the total costs and the impact of the costs in RIY) and the breakdown is now solely presented in the second table, which reduces the volume of information included in the cost section of the KID. General comments on the list of costs of investment funds

163
The consultation paper did not contain specific questions on the list of costs of investment funds. Feedback provided in the general comments of stakeholders focused in particular on the four following cost elements:
Loading costs: Several respondents asked whether these costs are identical to subscription fees. The ESAs deleted the term loading costs, while they broadened the category of subscription fees to ensure that no element linked to this category would be excluded. Providers of property management services for real estate operating expenses and capital expenditure: Some stakeholders questioned the scope of this cost item. However, they provided for diverging views as to whether capital expenditure should include expenditures linked to repair, renewal and replacement or the construction of new elements on the property. In addition, one respondent considered that only fees to persons for outsourced services are covered but not operating expenses and effective capital expenditure costs. The ESAs consider that the level of detail discussed extends the scope of level 2 work, and wants to avoid a situation in which a specific term could lead to an unjustified exclusion of certain elements. Carried interests: One stakeholder clarified that carried interests are not typically paid each year. In many cases, these have to be paid in the disinvestment phase, once the external investors have received their drawn down capital plus an agreed return. The ESAs concluded that those carried interests which are paid throughout the recommended holding period must be included in the cost calculations, and those carried interests which are paid afterwards must only be disclosed. Look-through approach: Several stakeholders noted that the wording regarding the costs of underlying investments is ambiguous. The ESAs clarified the wording in such a way that the different steps for calculation are now precisely outlined. Reduction in Yield (RIY) and other issues related to the costs of structured products Question 25 In relation to paragraph 68 a) of Annex VI: Shall the RTS specify that for structured products calculations for the cost free scenario have always to be based on an adjustment of the payments by the investor? Although the consultation paper did not include specific questions on structured products, respondents from the industry provided feedback on the presentation of costs of these types of PRIIPs. In particular, the structured products associations supported the annualised presentation arguing that it is the most frequently used format for cost disclosure to retail investors in the financial industry. One pointed out that the annualized amounts could be artificial for structured products with upfront/embedded costs and no annual/ongoing fees. Another one expressed a preference for the arithmetic average instead of the discounting. Regarding the Tables on Annex VII respondents from the structured products industry supported the percentage format in Table 2 unless the evidence shows that retail customers do not understand percentage figures. Some of them suggested adding that all costs in % in the second
table are expressed per year so their sums add up to the RIY. They do not see difficulties in
calculating the monetary value in Table 1.

164
Many respondents (not only from the SP industry) expressed strong concerns about the breakdown of costs in Table 1 which is, in their opinion, not requested at L1; only a few were in favour of a detailed presentation of costs. Almost all respondents argued that for structured products calculations for the cost free scenario have always to be based on an adjustment of the payments by the investor, as investors in structured products often have no possibility to make additional investments after the initial subscription. Impact Assessment Question 28 Do you have any comments on the problem definition provided in the Impact Assessment? Are the policy issues that have been highlighted, in your view, the correct ones? If not, what issues would you highlight? Do you have any views on the identified benefits and costs associated with each policy option? Is there data or evidence on the highlighted impacts that you believe needs to be taken into account? Do you have any views on the possible impacts for providers of underlying investments for multioption products, and in particular indirect impacts for manufacturers of underlying investments used by these products, including where these manufacturers benefit from the arrangements foreseen until the end of 2019 under Article 32 of the PRIIPs Regulation? Are there significant impacts you are aware of that have not been addressed in the Impact Assessment? Please provide data on their scale and extent as far as possible. Only a few stakeholders reacted to the problem definition and the policy issues as the basis for the impact assessment. This question is used by most respondents to raise their overall concerns with the KID, the implementation process and the practical functioning. The ESAs tried to address these issues already under ‘general statements’ at the start of this feedback statement. On the actual impact assessment, a number of stakeholders would like to see more practical evidence of how chose methodologies work out in practice, i.e. what they would mean for different products. The ESAs included examples of these practical effects and extracts of the tests they did on a broad scale of PRIIPs to the Impact Assessment that is also enclosed in this package.

10 November 2015 165
EIOPA
Insurance and Reinsurance Stakeholder Group
Reply form for the
Consultation Paper on PRIIPs Key Information Documents

166
Question 1
Would you see merit in the ESAs clarifying further the criteria set out in Recital 18 mentioned above by way of guidelines? The comprehension alert would lose its value and would not help retail investors if it was used for a wide range of products including some that should not fall under its scope. At this stage, the criteria set out in recital 18 do not provide for complete legal certainty as to which products’ KIDs should bear such a comprehension alert. There may be, therefore, a risk that some PRIIPs manufacturers face a situation where they choose to disclose the comprehension alert although their products should not bear it. Therefore we are strongly in favour of a further clarification of these criteria. However insurance-based investment products differ significantly between Member states. Therefore some members consider that it might be more adequate to have the guidelines drawn up by the respective NSAs. Question 2 (iii) Would you agree with the assumptions used for the proposed default amounts? Are you of the opinion that these prescribed amounts should be amended? If yes, how and why? (iv) Would you favour an approach in which the prescribed standardised amount is the default option, unless the PRIIP has a known required investment amount and price which can be used instead? Some members consider that defining prescribed standardised amounts is important in order to allow the comparison between products throughout Europe. Notwithstanding the above, there are other members that are of the opinion that fine-tuning or detailing the assumptions in the regulatory technical standards (RTS) at EU level (such as setting the initial amount invested) might sometimes prove to be very difficult notably because of (1) the different spectrums of products available in different markets and (2) the differences in investment behaviour and capital at expense across the EU. This fine-tuning should be in line with consumer behaviour at national level. Question 3 For PRIIPs that fall into category II and for which the Cornish Fisher expansion is used as a methodology to compute the VaR equivalent Volatility do you think a bootstrapping approach should be used instead? Please explain the reasons for your opinion? The methodology under category III (involving bootstrapping and a minimum of 10.000 forwardlooking simulations) is much more complicated than the one under category II (historical volatility over the last 5 years). Instead of applying an excessively complex methodology to category II as well, the methodology behind category III products should be simplified. One member considers that a common approach for all product categories would be preferable in order to ensure the comparability of the results which is an important goal of level 1 of the regulation. This member is of the opinion that a forward looking simulation can be used for all products. Cornish-Fischer and bootstrapping each only deliver meaningful results for a small
section of the product scope. The backward looking nature of the approaches invites gaming.
Products could be optimized for the recent history whereas the future might holder a wider variety of scenarios. Question 4 Would you favour a different confidence interval to compute the VaR? If so, please explain which confidence interval you would use and state your reasons why.

167
A risk measure that regards to more than one point of the distribution would be preferred. Looking at a single point is only significant if one is looking at the normal distribution. Question 5 Are you of the view that the existence of a compensation or guarantee scheme should be taken into account in the credit risk assessment of a PRIIP? And if you agree, how would you propose to do so? Some of the members consider that the existence of a compensation or guarantee scheme should not be taken into account in the credit risk assessment of a PRIIP because these schemes are too diverse and it could be difficult for consumers to assess the quality of the guarantee (ex-ante fully funded schemes vs ex-post funded schemes). Additionally, in many cases the scheme is only providing a guarantee up to a certain amount. To take into account such schemes, the KID may well mention narratively whether the PRIIP is covered by a guarantee scheme and which one. Having said that though, there are other members that consider that it is a reality that from the perspective of the customer insurers´ credit risk in some countries is lower thanks to insurance guarantee schemes which should, therefore be taken into account as a mitigating factor when assessing credit risk. Question 6 Would you favour PRIIP manufacturers having the option to voluntarily increase the disclosed SRI? In which circumstances? Would such an approach entail unintended consequences? The aim of designing a meaningful SRI methodology should be that there is no need for changing the outcome of the calculation. Question 7 Do you agree with an adjustment of the credit risk for the tenor, and how would you propose to make such an adjustment? Some members agree with the proposal to adjust credit risk for the tenor, because credit risk is usually increasing with the tenor. Notwithstanding the above, other members consider that it is not appropriate, in order to avoid complexity, to discriminate further with an adjustment of the credit risk to take into account the applicable tenor. Question 8 Do you agree with the scales of the classes MRM, CRM and SRI? If not, please specify your alternative proposal and include your reasoning. No. In spite of the EU initiatives in progress to reduce the overreliance on credit ratings issued by credit rating agencies, it is surprising to see that the proposal considers that no suitable methodology other that the external rating has been found regarding credit risk assessment. It should be reminded that hundreds of insurance-based investment products´ manufacturers don´t have a credit rating issued by a credit rating agency. It should also be reminded that the cost of getting a rating by a credit rating agency is very high, especially for small and medium insurance undertakings, something that could distort competition.

168
The proposal provides that if an insurer doesn’t have a rating, it will be automatically allocated to CR3. According to the aggregation method proposed, this would mean that even if the insurer has a MR1, its final SRI would be 3. This is not a reasonable outcome, given that market risk is the most relevant factor for insurance-based investment products. For insurance-based investment products, the market risk should be factored in in a much more prominent manner than what is proposed in the current draft RTS. The ESAs’ alternative scale proposed page 9 is already a step in the right direction. It should also be pointed out that the MRM and SRI scales are not really appropriate for long term PRIIPs as they penalise broad equity based PRIIPs versus fixed income ones, while on the long term, equity products might offer better protection against inflation and be less volatile than fixed income based PRIIPs. Question 9 Are you of the opinion that for PRIIPs that offer a capital protection during their whole lifespan and can be redeemed against their initial investment at any time over the life of the PRIIP a qualitatively assessment and automatic allocation to MRM class 1 should be permitted? Are you of the opinion that the criteria of the 5 year tenor is relevant, irrespective of the redemption characteristics? Some members are of the opinion that a nominal capital protection at maturity should not mean an automatic allocation of the product to MRM class 1, because this would mean not taking into account the effect of inflation, especially for the longer tenors. However, there are other members that are of the opinion that for PRIIPs that offer a capital protection at maturity, a qualitative assessment and automatic allocation to MRM class 1 should be permitted regardless of their tenor. These members consider that the impact of inflation on the value of the PRIIP should not affect the market risk mainly because inflation is not a risk inherent 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 (competition issue). Additionally, it should also be taken into account the current low inflation rates and the expected prolonged low interest rates environment when assessing this particular issue. Question 10 Are you aware of other circumstances in which the credit risk assessment should be assumed to be mitigated? If so, please explain why and to what degree it should be assumed to be mitigated? Some of the members consider that the existence of a compensation or guarantee scheme should not be taken into account as a mitigating factor in the credit risk assessment of a PRIIP because these schemes are too diverse and it could be difficult for consumers to assess the quality of the guarantee (ex-ante fully funded schemes vs ex-post funded schemes). Additionally, in many cases the scheme is only providing a guarantee up to a certain amount.
Having said that though, there are other members that consider that it is a reality that from the perspective of the customer insurers´ credit risk in some countries is lower thanks to insurance guarantee schemes which should, therefore be taken into account as a mitigating factor when assessing credit risk. These members consider that the credit risk assessment should also be assumed to be mitigated because of insurance claims take precedence over other claims against

169 the insurance undertaking (subordination of claims provided in Article 275(1)(1) of Solvency II Directive). Question 11 Do you think that the look through approach to the assessment of credit risk for a PRIIP packaged into another PRIIP is appropriate? Yes, it is. Question 13 Are you of the opinion that the current Consultation Paper sufficiently addresses this issue? Do you it is made sufficiently clear that the value of a PRIIP could be significantly less compared to the guaranteed value during the life of the PRIIP? Several alternatives are analysed in the Impact Assessment under policy option 5: do you see any additional analysis for these assessment? The PRIIP Regulation establishes that there is a single risk indicator (Article 8(3)(d)). This is to be a summary indicator, which takes account of and combines the relevant factors. Thus, the presentation of several risk indicators for different intermediate stages as suggested in option 5.2 would be contrary to the level 1 text. Its limitations should be also explained. In our view, a warning, specifying the boundaries of the risk indicator would make sense. Furthermore, the PRIIP Regulation foresees an entire section of the KID for the description of what happens if consumers take out money early (Article 8(3)(g)(iv)). Thus, consumers are informed in this section about what happens when they surrender early. If the same information is included differently in different sections, this would only lead to confusion. The same applies to the option 5.1. Consumers will be confused if the term for the risk indicator was shorter than the term of the product displayed in Article 8(3)(d). Moreover, a risk indicator based on short and standardised holding period for all products is not meaningful since for longterm products, such as insurance-based investment products, consumers will receive a wrong impression about the real risk of the product. Question 18 Do you agree that the monetary values indicated in the first table are a sum of costs over the respective holding periods? Or should the values reflect annualized amounts? If you prefer annualized amounts, which method for annualisation should be used (e.g. arithmetic average or methods that consider discounting effects)? Some members consider that the monetary values indicated in the first table should be a sum of costs over the respective holding periods. However, there are other members that are of the opinion that since the insurance-based investment products have terms that sometimes last over decades, only annualised costs are comparable for different PRIIPs in a consistent, robust and stable way. If the total costs were included, then a product with a longer term would automatically – even if it is cheaper – look more expensive than a product with a shorter term. An option presenting the total costs for the whole investment period would not allow for an effective comparison between, for example, a product with a few months investment
period and one characterised by a 35 years investment and would be nothing but misleading for consumers.

170
Question 19
Do you think that estimating the fair value of biometric risk premiums as stated in paragraph 55(b) of Annex VI would raise any technical or practical difficulties? In order to achieve meaningful comparisons between products, the biometric risk premium and the investment costs cannot be aggregated in one figure and must be presented in separate sections of the KID. The PRIIPs Regulation is important to help enhance consumer protection and improve consumer confidence by aiming to improve the transparency and comparability of PRIIPs products. It is, therefore, extremely important that the features of insurance-based investment products are appropriately presented in the key information document. The ESAs acknowledge that the aggregation of the investment costs and the full biometric risk premium would be inappropriate is welcomed. It is indeed, the insurance sector’s views that, such an aggregation would (1) not seem to be in line with the level 1 PRIIPs Regulation; (2) not be in the interest of consumers who will not be in a position to compare what is comparable; and (3) create an unlevel playing field for insurance-based investment products. Nevertheless, only if the full biometric risk premium is presented separately, would consumers actually be able to make a meaningful comparisons. Meaningful comparison remains the key objective of the PRIIPs Regulation and the insurance sector considers that only separating the full biometric risk premium from the investment costs could achieve such an objective. If this separation is not made, the consumer will be disadvantaged in several ways, as they would not be in a position to compare what is comparable:
 The cost indicator of an insurance-based investment product will be deceptively higher than that of other PRIIPs, and consumers will not be in a position to compare the investment part of the different products on the market.  The amount of the insurance premium will not be clearly visible to consumers and this will prevent them from comparing the insurance cover, including the potentially high benefits if the insurance cover payment is granted. It will also allow them to compare the premium with the ones offered through other insurance-based investment products and through pure life insurance products with no investment component. Therefore, in order to achieve meaningful comparisons between products, these two features cannot be aggregated in one figure and must be presented in separate sections of the KID. Question 20 Knowing that the cost element of the biometric risk premium is included in the total costs calculation, how do you think the investor might be most efficiently informed about the other part of the biometric risk premium (i.e. the fair value), and/or the size of biometric risk premium overall? Do you consider it useful to include the fair value in a separate line in the first table, potentially below the RIY? Or should information on the fair value be disclosed in another part of the KID (for instance, the “What is this product?” section, where the draft RTS currently disclose biometric risk premiums in total, and/or in the performance section)? What accompanying narrative text do you think is needed, and where should this be placed, including specifically narrative text in the cost section?

171
First, it seems key to recall that the level 1 PRIIPs Regulation Article 8(f) introduces in the KID a
section on costs which should include “the costs associated with an investment in the PRIIP” – it
does not say “costs associated with an investment and biometric protection”. Therefore, separating the full biometric risk premium and the investment cost, as well as being the most transparent and meaningful approach, is also in line with the level 1 text. Insurance-based investment products comprise an insurance cover, consisting of protection against biometric risks faced by consumers, alongside an investment element. When freely choosing an insurance-based investment product, a consumer is looking for both beneficial investment opportunities and for insurance protection for his or her family against biometric risks. Effective comparison should be ensured for consumers. Meaningful comparison remains the key objective of the PRIIPs Regulation and the insurance sector considers that only separating the full biometric risk premium from the investment costs could achieve such an objective. It is in the interest of the consumer that:
 The biometric risk premium in total is presented in a section separate from the KID cost section  No part of the insurance biometric risk premium is presented in the cost section of the KID  To ensure complete transparency, a reference to this could be made in the cost section, such as:
"The contributions for additional benefits that are not related to the savings process are presented separately.” Similarly, a reference to this separate section could be made in the performance scenario section, such as: "The additional benefits that are not related to the savings process are presented separately." It is not considered it useful to include the fair value component of any biometric risk premium in the cost section below the RIY. If optional, the biometric risk coverage should be priced separately as a non-mandatory option. If it is, on the contrary, embedded in the product, then there should be a narrative explanation included in the section “What is this product”. Question 21 Given evidence as to the difficulties consumers may have using percentage figures, would you prefer an alternative presentation of the second table, solely using monetary values instead? As with the first table, please also explain what difficulties you think might arise from calculating monetary values, and whether this should be on an annualized basis, and if so, how? The question relates to the second table on the presentation of costs. The ESAs proposed format of presentation of the costs is not supported. First of all, only the costs at the recommended holding period or at maturity are meaningful. Otherwise, not only would it lead to confusing information and information overload for consumers, but such a presentation would simply send the wrong message to retail investors:
 Insurance-based investment products are usually long-term products. These products are being considered by retail investors also for this very feature. When acquiring an insurance-based investment product, the retail investor should aim to keep it until the recommended holding period or at maturity. Displaying, at a pre-contractual stage, holding periods inferior to the recommended holding period or the product’s maturity would send the wrong message to retail investors.  Obviously, surrender costs are important and must be displayed in a fully transparent way to retail investors. It is understood that this is the reason why the PRIIPs level 1 Regulation dedicates a specific section of the KID to the surrender. The issue would be better addressed in the section of the KID on surrender value.  In addition, it should be pointed out that the RIY method, selected by the ESAs, has the advantage of taking into account the timing of costs, compared to the Total Cost Ratio. In this context, it seems irrelevant to display all these costs figures over time.

172
 Furthermore, it should also be pointed out that Level I regulation does not mention intermediate holding periods. Should, against all the above arguments, it be decided to include intermediate periods, it must be acknowledged that the draft RTS must consider the different products included in the PRIIPs Regulation, including long-term (such as insurance-based investment products) and very short term investment products. Regarding insurance-based investment products specifically, it should be noted that adding scenarios for intermediate stages of 1, 3 and 5 years as suggested in Annex VII makes no sense, given that insurance-based investments generally have very long recommending holding periods (of 30 years and more in many jurisdictions). Question 26 Regarding the first table of the cost section presented in Annex VII, would you favour a detailed presentation of the different types of costs, as suggested in the Annex, including a split between one-off, recurring and incidental costs? Alternatively, would you favour a shorter presentation of costs showing only the total costs and the RIY? We favour a shorter presentation of costs showing only the total costs and the RIY. Question 27 Regarding the second table of the cost section presented in Annex VII, would you favour a presentation of the different types of costs showing RIY figures, as suggested in the Annex, or would you favour a presentation of costs under which each type of costs line would be expressed differently, and not as a RIY figure -expressed as a percentage of the initial invested amount, NAV, etc.? We do not favour the second table, which is unnecessary. But if the second table is to be kept, it is important that costs are expressed using in the same approach in the two tables (RIY figures). It could be very confusing for investors if two different units (percentage of the initial invested amount, NAV) are used.

173
EBA
Banking Stakeholder Group
Reply form for the
Consultation Paper on PRIIPs Key Information Documents

174
London, 29 January 2016
Foreword
The EBA Banking Stakeholder Group (BSG) welcomes the opportunity to comment on the Joint Consultation Paper JC//2015/073 on PRIIPS Key Information Documents. This response has been prepared on the basis of comments circulated and shared among the BSG members and members of the BSG’s Technical Working Group on Consumer Protection and Financial Innovation. The response outlines general comments by the BSG, as well as detailed answers to the questions indicated in the Consultation Paper. General comments A critical aspect in retail distribution of any financial product is the transparency of product information. This represents not only the availability of all product information to retail clients but also the appropriate form of presenting the information to retail investors in the form easy to understand and enabling these investors to comprehend key product features and associated risks and costs. This is particularly important in respect of investment products and financial products (instruments) whose financial performance depends on uncertain and an often complex set of factors where significant risk of investor detriment exists. 14 The aim of regulation regarding disclosure under PRIIPs Regulation is twofold - in addition to enhancing transparency of packaged retail and insurance-based investment products it aims at setting up a harmonised template for comprehensive disclosure of information about PRIIPs which enables comparability of individual PRIIPs. This is designed to enhance informed choice for retail investors. This inevitably leads to careful consideration of complex issues related to comparability and comprehensibility of information which should be provided to retail investors in respect of individual types of PRIIPs. BSG appreciates the highly demanding task of ESAs to develop suitable templates and methodology covering very diverse groups of investment and insurance products and aims at provide the ESAs with its contribution to their effort enhancing the standard of PRIIPs disclosure to retail investors. Responses to the consultation questions Question 1 Would you see merit in the ESAs clarifying further the criteria set out in Recital 18 mentioned above by way of guidelines? Some of the criteria included in Recital 18 of PRIIPs Regulation 15 represent terms of a general nature and do not refer to specific legal terms which could lead not only to diverse

14 For example cf. Moloney N., How to Protect Investors. Lessons for the EC and the UK. Cambridge University Press.
2010. p. 191 et seq.
15 "[...] A product should be regarded as not being simple and as being difficult to understand in particular if it invests in underlying assets in which retail investors do not commonly invest, if it uses a number of different mechanisms to calculate the final return of the investment, creating a greater risk of misunderstanding on the part of the retail investor

175 implementation in member states but also to confusion for retail investors and hinder comparability of KIDs. Strict interpretation could lead to the conclusion that most, if not all, PRIIPs represent complex products and vice versa. In this respect further guidance provided by the ESAs is welcomed. Question 2 (v) Would you agree with the assumptions used for the proposed default amounts? Are you of the opinion that these prescribed amounts should be amended? If yes, how and why? (vi) Would you favour an approach in which the prescribed standardised amount is the default option, unless the PRIIP has a known required investment amount and price which can be used instead? Setting similar assumptions for all products would most likely result in retail investors not receiving relevant information and certain products outperforming others based on the KID although they might not be the best fit for all retail investors. In practice, actual amounts retail investors would use for comparison of PRIIPs could differ significantly among member states. Also retail investors in member states outside the EURO zone may find it difficult to compare PRIIPs using EURO denominated default amount. RTS should not lead to the provision of information that might be confusing or even misleading to retail investors. Therefore, fine-tuning should be in line with consumer behaviour at national level. Possible solutions could include setting a harmonized single default amount in EUR and provide an option for member states to set an additional default amount at national level in national currency where applicable. Question 3 For PRIIPs that fall into category II and for which the Cornish Fisher expansion is used as a methodology to compute the VaR equivalent Volatility do you think a bootstrapping approach should be used instead? Please explain the reasons for your opinion? The methodology under category III (involving bootstrapping and a minimum of 10.000 forwardlooking simulations) is much more complicated than the one under category II (historical volatility over the last 5 years). We understand that bootstrapping used for category III and IV is used as a substitute in case historical data are not available. Instead of applying an excessively complex methodology to category II as well, the methodology behind category III products could be simplified. Question 4 Would you favour a different confidence interval to compute the VaR? If so, please explain which confidence interval you would use and state your reasons why. Yes, we agree with the suggested confidence interval. Question 5 Are you of the view that the existence of a compensation or guarantee scheme should be taken into account in the credit risk assessment of a PRIIP? And if you agree, how would you propose to do so? The existence of a guarantee scheme could modify the level of risk as it potentially reduces the losses for relevant retail investors.

or if the investment's pay-off takes advantage of retail investor's behavioural biases, such as a teaser rate followed by a much higher floating conditional rate, or an iterative formula."

176
National compensation and guarantee schemes could be important components of the credit risk profile and hence these cannot be ignored in the CRM. A reference under the section “What happens if [PRIIP manufacturer] is unable to pay out?” does not provide sufficient information on credit risk. However, provision of additional information about the specific effects of such a scheme on an individual investor as part of the investor information/advice outside KID could be recommended. The credit risk retail investors could be facing, when purchasing a PRIIP, is mostly the risk linked to a PRIIP manufacturer’s insolvency. If this risk is already mitigated by a guarantee scheme, then it may become immaterial for the retail investor. Therefore, if the PRIIP manufacturer is fully secured through a guarantee scheme, then the credit risk from the point of view of retail investor is immaterial and the PRIIP should be categorised CR1. Question 6 Would you favour PRIIP manufacturers having the option to voluntarily increase the disclosed SRI? In which circumstances? Would such an approach entail unintended consequences? The main purpose of KID for PRIIPs is to enhance comparability of PRIIPs thus introducing the discretionary option to voluntarily increase the disclosed SRI which would lead to more difficult comparison of similar types of PRIIPs. Question 7 Do you agree with an adjustment of the credit risk for the tenor, and how would you propose to make such an adjustment? Adjustment for the tenor could make the methodology more complex. However, as taking into account the tenor in the credit risk assessment could be considered the market practice (also applied in credit rating methodologies) a similar adjustment to tenor could be envisaged. Question 8 Do you agree with the scales of the classes MRM, CRM and SRI? If not, please specify your alternative proposal and include your reasoning. Regarding MRM indicator:
The criteria in Category I “up to 5 years”, seems an artificial barrier, as explained in Question 9. The general comment regarding derivatives for hedging purposes, done in the “Introduction” is especially relevant here. Regarding CRM:
It may be questionable to base the credit risk assessment only on external ratings as a number of entities, particularly smaller entities and insurance-based investment products´ manufacturers, do not have a credit rating issued by an external credit rating agency. Also, cost implications of obtaining ratings for smaller entities are not negligible. Regarding SRI:
A SRI is not the best approach for the analysis of risks. As credit and market risk are not comparable, a separate analysis of both risks is recommended.

177
Question 9
Are you of the opinion that for PRIIPs that offer a capital protection during their whole lifespan and can be redeemed against their initial investment at any time over the life of the PRIIP a qualitatively assessment and automatic allocation to MRM class 1 should be permitted? Are you of the opinion that the criteria of the 5 year tenor is relevant, irrespective of the redemption characteristics? The allocation in class 1 for PRIIPS with capital protection during their whole lifespan and can be redeemed against their initial investment at any time over its life seems appropriate. As stated in the answer to Question 8, the five years tenor criteria seems arbitrary, irrelevant and based on no specific evidence. A guaranteed product with a maturity of more than five years does not have a higher market risk than a similar product with a shorter tenor. This arbitrary 5 year cap could also mislead retail investors. For PRIIPs that offer a capital protection at maturity, a qualitative assessment and automatic allocation to MRM class 1 should be permitted regardless of their tenor. The criteria of the 5-year tenor is justified in the document to address inflation concerns. The impact of inflation on the value of a PRIIP should not affect the market risk as inflation does not represent a risk inherent for PRIIPs but affects all investment products in the same way. In addition, this feature is not included in pre-contractual information disclosure for other products (MiFID and UCITS for instance) and could add unnecessary complexity. This distinction is not helpful, therefore, for retail investors nor does it increase comparability or transparency of products. Question 10 Are you aware of other circumstances in which the credit risk assessment should be assumed to be mitigated? If so, please explain why and to what degree it should be assumed to be mitigated? We consider that the existence of a compensation or guarantee scheme should not be taken into account as a mitigating factor in the credit risk assessment of a PRIIP because these schemes are diverse and it could be difficult for retail investors to assess the quality of the guarantee (ex-ante fully funded schemes vs. ex-post funded schemes, level of protection, eligibility of individual retail investor etc.). In addition, mostly such compensation and guarantee schemes provide compensation and/or guarantee only to a certain amount. Question 11 Do you think that the look through approach to the assessment of credit risk for a PRIIP packaged into another PRIIP is appropriate? In practice, it may be challenging to apply the look through approach for the complex PRIIPs. In order to make this approach generally feasible, all relevant data and explanations have to be provided by manufactures of underlying PRIIP. Question 12 Do you think the risk indicator should take into account currency risk when there is a difference between the currency of the PRIIP and the national currency of the investor targeted by
the PRIIP manufacturer, even though this risk is not intrinsic to the PRIIP itself, but relates to the typical situation of the targeted investor?

178
Currency risk should be taken into account if it is part of the PRIIP market risk - typically for structured products - if it can materially alter the value of the guarantee if given in another currency. In the case where the PRIIP is denominated in a currency that is not the currency of the retail investor domicile, then additional explanation should be provided by a distributor outside the KID. Question 13 Are you of the opinion that the current Consultation Paper sufficiently addresses this issue? Do you it is made sufficiently clear that the value of a PRIIP could be significantly less compared to the guaranteed value during the life of the PRIIP? Several alternatives are analysed in the Impact Assessment under policy option 5: do you see any additional analysis for these assessment? The PRIIPs Regulation sets out a specific section of the KID dedicated to the description of what happens if consumers take out money early. 16 Thus, retail investors are informed in this section about specific consequences if the investor exits/sell the PRIIP before maturity. We are of the opinion that this information should not be duplicated in a KID as this may lead to confusion. Question 14 Do you agree to use the performance fee, as prescribed in the cost section, as a basis for the calculations in the performance section (i.e. calculate the return of the benchmark for the moderate scenario in such a way that the return generates the performance fee as prescribed in the cost section)? Do you agree the same benchmark return should be used for calculating performance fees for the unfavourable and favourable scenarios, or would you propose another approach, for instance automatically setting the performance fees to zero for the unfavourable scenario? Please justify your proposal. All scenarios should use the performance fee related to returns under relevant scenario - the fee applicable for the scenario should be used in calculation of return under such scenario. In the case where fees are not included in the calculation the information for the retail investor is incomplete and could lead to confusion and typically to investor detriment as the real outcome does not match the model outcome under relevant scenario. Question 15 Given the number of tables displayed in the KID and the to a degree mixed consumer testing results on whether presentation of performance scenarios as a table or a graph would be most effective, do you think a presentation of the performance scenarios in the form of a graph should be preferred, or both a table and a graph? The addition of a recommended holding period in itself adds complexity to any performance presentation. In practice retail investors (consumers) find it difficult to assess the period for which they expect to hold an investment. Nevertheless, the visualisation of performance scenario in the form of a graph could possibly provide retail investors with an easier—to-understand way of presenting complex structured data.
Question 16

16 In art. 8 par (3) let. (g) point (iv) requires manufacturer to include in a KID "information about the potential consequences of cashing in before the end of the term or recommended holding period, such as the loss of capital protection or additional contingent fees;".

179
Do you agree with the scope of the assets mentioned in paragraph 25 of Annex VI on transaction costs for which this methodology is prescribed? If not, what alternative scope would you recommend? In the costs section the alignment with MiFID 2 is crucial, bearing in mind that while KID discloses manufactures’ costs, MiFID 2 regulates disclosures of costs at the level of the firm providing investment services to retail investor such as selling the PRIIP. Regarding costs in Annex VI: Spread is not a cost. Presenting a spread as a cost may create an impression that it is possible to conclude a trade at mid-price. The bid/mid spread is a measure to mitigate credit risk, that the bank would be imposed by the trade, as well as the cost of hedging the position. Question 17 Do you agree with the values of the figures included in this table? If not, which values would you suggest? (please note that this table could as well be included in guidelines, to allow for more flexibility in the revision of the figures) No comment Question 18 Do you agree that the monetary values indicated in the first table are a sum of costs over the respective holding periods? Or should the values reflect annualized amounts? If you prefer annualized amounts, which method for annualisation should be used (e.g. arithmetic average or methods that consider discounting effects)? Since some PRIIPs (especially insurance-based investment products) have terms that sometimes stretch over decades, only annualised costs are comparable for different PRIIPs in a consistent and stable manner. In the case where the total costs were to be included, a product with a longer term would typically – even if it is cheaper – look more expensive than a product with a shorter term. An option presenting the total costs for the whole investment period would not allow for an effective comparison between, for example, a product with a few months investment period and one characterised by a 35 years investment and would be nothing but misleading for retail investors. Question 19 Do you think that estimating the fair value of biometric risk premiums as stated in paragraph 55(b) of Annex VI would raise any technical or practical difficulties? In order to achieve meaningful comparisons between products, the biometric risk premium and the investment costs cannot be aggregated in one figure and must be presented in separate sections of the KID. The PRIIPs Regulation aims at enhancing retail investor protection and confidence by improving the transparency and comparability of PRIIPs. It is, therefore, extremely important that the features of insurance-based investment products are appropriately presented in the KID. We support the view of ESAs that the aggregation of the investment costs and the full biometric risk premium would be inappropriate. It is indeed the insurance sector’s views that such an aggregation would (1) not seem to be in line with the PRIIPs Regulation; (2) would not be in the

180 interest of consumers who will not be in a position to compare what is comparable; and (3) would create an unlevel playing field for insurance-based investment products. If this separation is not made, the consumer will not be provided with appropriate information and would not be in a position to compare what is comparable:
The cost indicator of an insurance-based investment product will be deceptively higher than that of other PRIIPs, and retail investors (especially consumers) will not be in a position to compare the investment part of different insurance-based investment products available on the market. The amount of the insurance premium will not be clearly visible to consumers and this will prevent them from comparing the insurance cover, including the potentially high benefits if the insurance cover payment is granted. Question 20 Knowing that the cost element of the biometric risk premium is included in the total costs calculation, how do you think the investor might be most efficiently informed about the other part of the biometric risk premium (i.e. the fair value), and/or the size of biometric risk premium overall? Do you consider it useful to include the fair value in a separate line in the first table, potentially below the RIY? Or should information on the fair value be disclosed in another part of the KID (for instance, the “What is this product?” section, where the draft RTS currently disclose biometric risk premiums in total, and/or in the performance section)? What accompanying narrative text do you think is needed, and where should this be placed, including specifically narrative text in the cost section? Insurance-based investment products comprise an insurance cover, consisting of protection against biometric risks faced by consumers, alongside an investment element. When freely choosing an insurance-based investment product, a consumer is looking for both beneficial investment opportunities and for insurance protection his or her family against biometric risks. The life insurance part of an insurance-based investment product may offer a number of benefits:
(a) Protection of surviving dependents: first and foremost, death benefits provide surviving family members with funds allowing them to maintain their living standards. For instance, it can provide funds for college education when the principal income earner is deceased and/or a financial safety net to offset the impact of estate taxes upon the policyholder’s death. (b) Income protection: benefits that ensure a stable living income in the event that the consumer is not able to exercise his/her profession or work in any capacity, either temporarily or permanently. (c) Succession planning: allows a customer to save or invest money for his/her children or grandchildren while keeping control over the funds and the time of pay-out (e. g. not automatically after a certain period of time). (d) Long-term care: the organisation and delivery of a broad range of services and assistance to people who become limited in their ability to function independently on a daily basis over an extended period of time due to mental and/or physical disability. (e) Consistent saving: compared to saving accounts, regular payments of a premium offers the consumer a more disciplined way of saving.

181
All these benefits are unique to insurance-based investment products and are secured by the payment of the insurance premium (i.e. the price to pay in exchange for these insurance services). A sharp and clear distinction must, therefore, be made between investment costs associated with the insurance-based investment product and the insurance premiums paid. Premiums — which are payments that directly finance the insurance benefits of the products — should never be considered as costs. This is simply because the consumer knowingly receives insurance benefits for these payments and in fact specifically chooses an insurance-based investment product in order to receive these benefits along with investment returns. If the consumer is not interested in receiving additional insurance benefits, he or she would not opt for an insurance-based investment product in the first place. However, if consumers are interested in receiving additional insurance benefits, the presentation of insurance premiums as investment costs would not give them the appropriate and necessary information on the product. Effective comparison should be ensured for consumers. Meaningful comparison remains the key objective of the PRIIPs Regulation and the insurance sector considers that only separating the full biometric risk premium from the investment costs could achieve such an objective. However in the case where the insurance-based investment product contains no or only minimal (negligible) protection against biometric risks, the nature of the product should be approximate to underlying investment and treated as a "regular" investment product. Question 21 Given evidence as to the difficulties consumers may have using percentage figures, would you prefer an alternative presentation of the second table, solely using monetary values instead? As with the first table, please also explain what difficulties you think might arise from calculating monetary values, and whether this should be on an annualized basis, and if so, how? Disclosure of the costs should be aligned with MiFID 2 and hence the table should include both percentage figures and monetary values. Question 22 Given the number of tables shown in the KID, do you think a more graphic presentation of the breakout table should be preferred? The example of a graphical breakdown presentation offered on page 14 of the consultation paper adds little in terms of clarity for clients compared to a table. Question 23 The example presented above includes a possible way of showing the variability of performance fees, by showing the level for all three performance scenarios in the KID, highlighting the ‘moderate‘ scenario, which would be used for the calculation of the total costs. Do you believe that this additional information should be included in the KID? Additional information regarding different performance fees should be included in the KID. By their nature performance fees are rather incidental than recurring. Information on
different fees could be included in the last row of the Table 2 - we suppose that these are the same as performance fees used for scenario calculations. Question 24

182
To reduce the volume of information, should the first and the second table of Annex VII be combined in one table? Should this be supplemented with a breakdown of costs as suggested in the graphic above? Reducing the volume of information should only be done if there is no lack of clarity. The combination should be done in a way that guarantees a better result than the separate tables. Question 25 In relation to paragraph 68 a) of Annex VI: Shall the RTS specify that for structured products calculations for the cost free scenario have always to be based on an adjustment of the payments by the investor? We agree to the proposed method. Question 26 Regarding the first table of the cost section presented in Annex VII, would you favour a detailed presentation of the different types of costs, as suggested in the Annex, including a split between one-off, recurring and incidental costs? Alternatively, would you favour a shorter presentation of costs showing only the total costs and the RIY? The breakdown of one-off, recurring and incidental costs could be seen as adding unnecessary complexity to the information presented in KID. This approach mixes defined and possible costs as well as distribution and product costs. Specifically, as for mentioning distribution costs in a KID, cost information in the KID should be limited to product costs, as distribution costs may vary depending on the party distributing the PRIIP. In which case, information on distribution cost could be more appropriately provided outside the KID. Question 27 Regarding the second table of the cost section presented in Annex VII, would you favour a presentation of the different types of costs showing RIY figures, as suggested in the Annex, or would you favour a presentation of costs under which each type of costs line would be expressed differently, and not as a RIY figure -expressed as a percentage of the initial invested amount, NAV, etc.? Expressing costs as a percentage of the initial invested amount is preferred as a clear and simple way of presenting relevant information. Question 28 Do you have any comments on the problem definition provided in the Impact Assessment? Are the policy issues that have been highlighted, in your view, the correct ones? If not, what issues would you highlight? Do you have any views on the identified benefits and costs associated with each policy option? Is there data or evidence on the highlighted impacts that you believe needs to be taken into account? Do you have any views on the possible impacts for providers of underlying investments for multioption products, and in particular indirect impacts for manufacturers of underlying investments used by these products, including where these manufacturers benefit from the arrangements foreseen until the end of 2019 under Article 32 of the PRIIPs Regulation?

183
Are there significant impacts you are aware of that have not been addressed in the Impact Assessment? Please provide data on their scale and extent as far as possible. The Impact Assessment included in the Consultation Paper is a very detailed, complete and structured study. However, regarding policy issues, two of them are missed:
The guidance on providing of the KID to the retail investor remains overly general in nature, especially when it comes to situations where there is no physical meeting between the distributor and the client. The criteria mentioned in art. 20 of the draft RTS relate solely to the timing of the delivery of the KID, and do not provide any guidance on any of the other practical issues of the provision of the KID and additional information to retail investors such as the manner in which the KID can be delivered to the retail investor (bundled with other KIDs distributor´s website, by means of a hyperlink or as an attachment to electronic communication, etc.). Further clarification should be provided specifying PRIIPs for which a KID must be prepared and distributed as of 1 January 2017 e. g. only for new products launched as of 1 January 2017 or products which will be launched before this date but could be subscribed/purchased by the retail investors on that date as well? Submitted on behalf of the BSG David T Llewellyn, Chairperson.

Lineage: No known changes

Similar documents from other regulators