In focus - Markets

Fixing the PRIIP Key Information Document

PRIIPs are supposed to be transparent and easy to understand, but owing to poorly designed regulation this is not the case. Here we look at some of the ways the regulation could be improved.

18/09/2019

Anastasia Petraki

Head of Policy Research

It is quite rare for investors to have strong reservations about regulation intending to help them with their decision making. But that was the fate of the new Key Information Document (KID) for Packaged Retail and Insurance-based Investment Products (PRIIPs). The intentions were good: improved disclosure and full comparability across different product types such as funds, pension, insurance and structured products. The results, less so:

  • a methodology that results in negative transaction costs
  • further obfuscation of cost information by presenting it in an unfamiliar way that deviates from what is actually paid
  • removing the past performance figures investors expect to see on the grounds that investors may incorrectly extrapolate it into the future, and replacing them with scenarios of theoretical future performance, which are based on that very same past performance

Recognising the issues, after intense negative commentary by both consumer groups and the fund management industry, policymakers have decided to review the regulation and there is currently a limited window until the end of the year to revisit the content of the Key Information Document.

This is the time for all stakeholders to work together constructively to ensure investors are not provided with misleading information that may steer them to wrong choices and poor outcomes. Perhaps most important of all, the consumer voice should be much more prominent in these discussions.

In theory, the first two issues above relating to cost disclosure can be resolved relatively easily, as set out in our paper on this subject. However, the practicalities of getting Europe-wide approval should not be underestimated.

The third is potentially more challenging. As things stand, past performance will be removed from all regulated disclosure materials and, instead, investors are going to be confronted with four scenarios: favourable, moderate, unfavourable, and stressed. These have been designed to help investors understand what they are likely to get in return if they stay with any given product over its recommended holding period.

Given the widely known caveat that past performance is no guide to the future, it is rather unfortunate that these scenarios will be based on the performance of the past five years. It is easy to see the shortcomings of such an inherently pro-cyclical approach. Following a period of very strong performance, even returns in the unfavourable scenario may be strongly positive. Conversely, after a period of strongly negative returns, such as the bursting of the Dotcom bubble or the Great Financial Crisis, the scenarios will predict more of the same. Rather than helping to predict future returns, our back-testing of the scenario methodology found a strong negative relationship between predicted returns and what was realised.

If it is not changed, the current approach will incentivise investors to buy when markets are high and sell when markets are low. This is the exact opposite of what they should be doing.

However, proposals made so far, such as basing the scenarios on the last ten instead of five years, fail to address the problem. So what would it take to fix this?

There is no silver bullet but, in this paper we explore five different options with a focus on what this would mean for consumers – our clients – and their decision making.

Option 1: The ‘hard way’ would be to allow for different performance disclosure for different types of products, allowing funds to show past performance while structured and some insurance products, for which there is no meaningful way to show past performance, would still show scenarios. This would address concerns around misleading performance information and would not affect comparability across products of the same type nor comparability of the other key information across products of different types. However, it would require changes to the overarching regulatory framework that would difficult to re-open and it would affect comparability of performance across different types of products.

Option 2: The ‘half way’ would be to simply show past performance alongside scenarios. Past performance is factual and important for accountability for delivery against the investment objective. In addition, this approach would not affect comparability. However, it would also have the potential to reinforce procyclical investment behaviour e.g. after a bull market, investors would see both very positive scenarios and high historical returns.

Option 3: The ‘factual way’ of fixing this would be to use past performance as a scenario. That is, find clever ways to show a range of potential outcomes based on what would have happened if investors had bought and sold their holdings at different points in time in the last ten years. See below for an example. Standard past performance figures could be shown in addition.

Illustrative-scenarios-potential-past-returns.jpg

This would have the advantage of showing factual information without making a forecast of what investors could expect in the future. But it would still, effectively, be based on past performance.

Option 4: The ‘asset class way’ would be to show fund-specific past performance alongside scenarios that relate to the fund’s asset class, proxied by the fund’s sector or benchmark. This would allow showing what could be a typical outcome for the asset class in which a fund has the highest exposure and have scenarios that are based on a period longer than the past five years. The problem would be that unless regulators come up with a new estimation methodology for the scenarios, the results would still be procyclical. It could also create additional complications in terms of defining a common framework for asset class and sector classification or finding solutions where funds cannot be classified in any one sector or have no benchmark.

Option 5: The ‘untested way’ would be to show past performance accompanied not by scenarios but by a narrative about the main drivers of performance and under what conditions investors could expect positive or negative outcomes. This could provide better context to the fund’s delivery in connection to its investment objective and help investors better understand the fund itself. The disadvantage of this approach, other than the fact that it hasn’t been tested with consumers, is that it would not involve a metric that is directly comparable across products. Moreover, large blocs of text may be off-putting for investors.

Two clear messages emerge from this.

First, anything resembling a forecast of returns has no place in consumer disclosure. Past performance is needed as a clear measure of delivery and accountability and we strongly believe that it should not be removed. Given that removal of scenarios entirely would be hardest to achieve from a regulatory standpoint, the transparent and intuitive nature of Option 3 has a lot going for it.

Second, although comparable information across a (very) diverse set of investment products is welcome in principle, comparability should not receive such primacy that it comes at the cost of providing meaningful information to consumers.

Read the full report

PRIIP Key Information Document 9 pages | 303 kb

DOWNLOAD