Investors may have noticed that super fund Product Disclosure Statements (PDSs) now include a measure of risk called the ‘Standard Risk Measure’, or SRM. The intention is to provide greater risk disclosure for retail investors. I encourage people to be very careful when reading such disclosures and to think about risk in more ways than is simply described by this measure in the PDS.
The SRM was introduced by the Australian Prudential Regulation Authority (APRA) in 2011. It is a self-assessed estimation by the product provider of the number of times over a 20-year period that a fund is expected to deliver negative returns. For example, an Australian equity fund might appear in a High Risk band because it is expected to generate negative returns in 5 out of every 20 years. However, a fixed income fund might be in a lower Medium Risk band because negative returns occur only once every 2.5 years. APRA sought market feedback prior to implementing the SRM but the original version was adopted unaltered. The proposal was supported by two industry groups which is an interesting story in itself that I discuss later.
Any effort to improve risk disclosure in retail PDSs is welcome. However, there has been considerable debate around whether this SRM represents a step forward or whether it creates a set of issues which exceed the benefits of greater risk disclosure.
I have many doubts about the SRM but focus on two in this article: the measurement itself and the calculation method.
There are many measures of risk in finance and no single risk measure is perfect. A mosaic of risk measures blended with experience and a qualitative appreciation is probably our best chance to understand risk. Each measure on its own provides useful information but is flawed. Using volatility alone assumes that we live in a world which is perfectly normally distributed. Using VaR (Value-at-Risk, an estimation of an adverse, statistically-rare outcome) effectively provides a data point around the loss in a rare event but leaves us with little knowledge about what will happen in an everyday environment.
Size of loss is more important than the frequency
There are two key elements to understanding risk: the size of an event and the likelihood of that event occurring. Consider how this applies to the SRM, where the size of an adverse event is ignored. An event is simplified to be any negative return. So a negative 5% return is not viewed any differently to a negative 50% return. Those approaching retirement prior to the GFC can vouch that a 50% negative has a major impact on their livelihoods. Indeed any investor would surely take three negative 5% return years rather than a single year of negative 50%, yet the SRM may in fact guide them to take the opposite position and only expect to lose money in one year.
There are some strategies which have a very low likelihood of loss but if they do lose, they lose substantially. Consider a fund for instance which sells out-of-the-money options. It may consistently make money year after year and then suddenly lose everything when an option is exercised. Such funds would quite correctly report a very low SRM, but I question if this is the outcome desired by APRA.
Too much subjectivity
The other main area of concern is implementation of the calculation, which is undertaken by the product provider, although APRA may review the calculation methodology. Even though there exists much science around calculating risk statistics, there remains much subjectivity. It is possible that two highly respected risk managers could look at an identical product and calculate a different SRM. And both calculations could be defended as having been calculated by a professional and backed with appropriate research.
This then creates a dilemma for product providers. Offered two different SRMs, there would be internal pressure to adopt the lower measurement, thereby making their product appear less risky and hence more attractive. There have already been some industry reports of similar products having different SRMs and of bond funds having a measure close to some equity funds.
I can empathise with APRA on this decision to have providers do their own calculations. Banks are required to calculate on a daily basis the amount of market risk they are taking (measured by VaR) which determines a capital requirement for market risk. Because the number of banks operating in Australia is relatively small, APRA is very hands-on in reviewing the calculation methodologies used by each bank. However, in the funds management industry, there are a huge number of products and fund providers and APRA has likely decided it is impossible to regulate this calculation closely.
But there were other implementation choices. A small team could have calculated the risk numbers, with the product provider given an opportunity to object. This may have led to greater consistency. Overall, the self-implementation approach reduces the ability of the SRM to be relied upon as a way of comparing products.
Lack of agreement between industry bodies
A particularly interesting aspect of the SRM debate has been the role of various industry bodies. Officially, the SRM is “the product of an ASFA (Association of Superannuation Funds of Australia) and FSC (Financial Services Council) working group, and is supported by ASIC (Australian Securities and Investments Commission) and Australian Prudential Regulation Authority (APRA).” However, the SRM was proposed by APRA prior to the creation of the working group. Many other industry bodies have criticised the statistic, notably AIST (Australian Institute of Superannuation Trustees), Actuaries Institute, and ISN (the Industry Super Network). It is confounding that these industry bodies can have such strongly opposed views, and perhaps leaves a question mark over the consultation process.
All up, while it is an admirable objective to improve risk disclosure, even given the difficulties of such a large universe of products and providers, I can only describe this as awkward regulation. I hope the Standard Risk Measure is not relied upon too heavily by retail investors.