The Key Questions are valid and they provide a potential objective basis for comparison across products. It would be desirable if they could also refer to the individual investor’s objectives. We would encourage ESAs to consider if it were possible for the KID to link the Key Questions to the broader types of outcome that would be of concern to investors with different risk appetites and objectives. In 2012 an IFoA Working Party published a discussion paper on transforming consumer information , in which one of the proposed outcomes was to “Communicate risk and reward by reference to the chances of achieving the consumer’s goal”.

The uncertainty of returns questions makes reference to ‘given current market conditions….’ If the assumptions underlying the performance scenarios are only refreshed periodically then this could introduce a comparability issue, since different providers may update their assumptions at different times and in different market conditions.

We also believe that the references to “winning” in the Key Questions (“How much can I win? How much am I likely to win?”) are not representative of how consumers relate to financial products. We suggest that “win” is replaced with “gain” or “earn”.

We agree that the definitions of market, credit and liquidity risk are quite broad, and therefore cover many of the key risks for PRIIPs. However, these broad definitions may be difficult to combine into a single quantitative measure. For example, the suggested measures of market risk are based on volatility and Value-at-Risk, both of which depend on historic or forecast returns. A single value of volatility, as an example, may be too crude to capture all the aspects of risk mentioned in the paper, such as inflation, interest rates, currency, and the investment’s underlying asset classes. [The IFoA’s paper on Transforming Consumer Information (see footnote 1) states (2.3.16): “if the consumer is saving for something in particular in 10 years, the price inflation of that particular item will have a significant bearing on the amount required. Indeed, the risk of long‐term returns being lower than expected is much more pertinent to success than the yearly volatility of the investment.”]

Many of the measures suggested are potentially calculation heavy and – more importantly – the end consumer is less likely to understand a large amount of technical language e.g. Value-at-Risk, credit spreads.

If the KID document is at product level rather than fund level, and the consumer’s investment choices are unknown, it may be more appropriate to use qualitative measures. The provision of context or benchmarking – as proposed by the visual examples in section 3.7 – are helpful, as without this, end consumers may struggle to draw conclusions from the information presented.

If the performance scenarios are to be based on forward looking assumptions, then in our view the risk measures should also be based on these, given the relationship between them.

If more than one scenario is to be included, then we would support having some indication of the likelihood of each.

Regarding performance scenarios in general, our view is the only way to achieve comparability between providers would be to apply a very prescriptive approach to the methodology for the scenarios. Such an approach is unlikely to succeed given the wide range of products and consumer objectives, and therefore there will be a trade off between comparability and flexibility of methods.

We would suggest that performance scenarios should use a combination of all these measures. In particular, providers should avoid using only percentages, since consumer research frequently indicates that consumers struggle with these.

We believe that gross performance figures are misleading as they can never be achieved. While there are challenges with presenting net of costs figures, they set more appropriate customer expectations.

Informing the customer must be the objective rather than simply providing information. We note the point that too few scenarios could trigger behavioural biases, but equally too many could impede effective decision-making. The optimum number will depend on the product and on the customer’s objectives.

The variety of examples shown in section 3.7 illustrates the benefits of enabling MS to design their own risk indicators. These will reflect the different products and frameworks within each MS. As noted already, for some insured investment products we would encourage consistency with the SRRI scale in KIIDs as consumers may be comparing alternative tax wrappers where KIIDs will be available for the underlying funds.

We support some form of benchmarking for the performance scenarios although this is potentially difficult to achieve. A balance needs to be struck between finding an engaging presentation that is simple enough for the majority of consumers to understand without it being misleading. For example, the ‘3 line graph’ on page 43 is potentially misleading as, without proper explanation, consumers may think that they could receive one of the straight lines.

The IFoA notes with interest the discussion of different presentation formats that may be used for summary risk indicators, performance scenarios and combinations of the two. This analysis is very useful in highlighting some of the principles that should guide the choice of presentation in practice. However, our view is that the most appropriate presentation format cannot be defined in advance, and that this is likely to vary for different member states. To achieve comparability between providers within each MS, the appropriate format would need to be prescriptively defined at an appropriate level.

As many of the costs are investment related and will therefore be dependent on an individual’s investment selection, we are unclear how this will be represented at the KID level. There may be a risk that product level disclosures are made based on the assumption of selecting the ‘cheapest’ investment option available.

We believe that the nature of indirect costs, particularly in structured products, will be almost impossible to bring together in a key document across the EU. While it is desirable for providers to strive to make such costs as transparent as possible to consumers, any attempt to standardise the description of these costs would inevitably cause customer communication difficulties.

If the customer knows the impact of costs e.g. a RIY percentage, in our view this is generally more useful than a detailed breakdown of the constituent parts. We would caution, however, that percentages are not generally well understood. Adopting an ongoing charge/TER approach could help improve comparability across different product types with different disclosure requirements e.g. KIIDs.

Transaction costs are not always readily identifiable. It would therefore be helpful to understand the extent to which the KID needs to recognise these costs. We would encourage the ESAs to have a general aim of consistency across product types and regulatory frameworks, both in terms of what is to be disclosed to consumers and also the methodology to be used by investment managers to calculate these costs. There needs to be clarity about the position of product providers who request the relevant information from investment managers but where this is not provided.

The products are fundamentally different so can have very different charging structures.

We believe that consumer testing is needed to inform this. Some of the examples provided are quite complicated.

We note the interesting discussion of the pros and cons of using a single indicator for risk in aggregate or separate indicators for each risk, and that the PRIIPs Regulation gives scope for some flexibility in this area. However, we would argue that it is mistaken to limit the KID to having a single form of presentation for risk. Instead, the five options identified in the Discussion Paper should be seen as available options for providers. Consumer testing could help them to choose appropriate risk indicators.

For some insured investment products, it would be helpful to have consistency with the SRRI scale in KIIDs, as consumers may be comparing alternative tax wrappers where KIIDs will be available for the underlying funds e.g. Offshore Bond vs Onshore Bond vs S&S ISA vs Pension.

We do not have specific comments to make about the costs and charges listed in table 12, but we would note in general that there should be a balance between providing sufficient transparency and causing confusion by providing too much information.