There is no one-size-fits-all solution concerning the most appropriate measure to evaluate risks as there are many different PRIIPs that need different measure to evaluate it. For instance, a risk measure appropriate for funds might be inappropriate for other types of PRIIPs.
As regards market risk, volatility is a well-known and well-established concept in finance, a measure conceptually easy to grasp and, at the same time, able to capture the effects of very different risk factors. We therefore support as a quantitative measure the historical ex-post volatility. However, for new issuances of products, this historical information is of course not available, but proxies could be available as an alternative.
We support that credit and liquidity risk should be better explained in qualitative format (narrative text).
In the interest of ensuring that the KID is usable for consumers, risk indicators should be as clear as possible for consumers. The market, credit and liquidity risks should thus not be aggregated. The different risks: market, credit and liquidity should be shown separately. Market risk could be explained with a quantitative measure most appropriate for the PRIIP in question. The other two in qualitative terms.
We therefore not support integrating all types of risk in a single indicator for the following reasons:
• Integrating all types of risk in a single indicator suggest at first sight the usage of some kind of risk aggregation methodology which, for the most part, entails methodological requirements we do not consider appropriate in order to gain insight and transparency for the investor. Instead, its apparent simplicity could be misleading and refrain the investor from a more critical and reflexive analysis in light of his personal circumstances making easier the appearance of cognitive biases.
• It is extremely important to note that the level of sensitivity every investor has in relation with each type of risk is in no way uniform. For some retail investors, liquidity could not be a major issue whilst credit risk losses could be unacceptable. A simple rule of thumb consisting in weighting arbitrarily some value for each type of risk would imply a set of given preferences that can be far away those of every individual investor.
• From a methodological perspective risk aggregation implies a set of pre-defined requirements. Namely, quantitative measures of loss variables for every type of risk, modeling dependencies between the different risks types which often implies the availability of historical data series and, last but not least, the need to overcome the fact that classical aggregation models operate in a single-period framework which is usually coincident with the planning horizon. Given the complexity of modeling with data not readily observable, comparability across products and issuers becomes a major issue.