Attempting to calculate risk through historic data can be an essential tool in providing an indication of future volatility
There are a number of comparative measurements of risk available to financial advisers when selecting suitable funds to recommend to clients, some of which are likely to become more widely used as regulatory and client demand for value increases.
In real terms, measurements of risk are all by necessity backward looking or based on historic data. However, if used carefully they can be used to provide something of an indication of future risk.
The most widely available of the measurements is the 12-month measurement of volatility as given by standard deviation. It is available in publications such as Investment Week's monthly sister publication Fund Strategies.
Standard deviation is a square root of the variance of absolute returns ' the dispersion of returns that a fund has earned over a period of time.
The rough guide is that the larger the number, the more volatile, or risky, a fund is. For example, according to Standard & Poor's, in the 12 months to 1 May, the average volatility in the UK All Companies sector was 4.1. That compared to 5.07 in the UK Smaller Companies sector, 1.43 in the UK Gilts sector and 10.44 in the Specialist sector.
That number, however, can hide a great deal of detail. It gives no indication of whether volatility was a positive or negative to performance but does provide a snapshot that can indicate how aggressive a fund is and how much absolute returns can vary.
Michael Owen, a director at investment intermediary group Plan Invest, said that at best volatility provides only one of a number of indicators of risk that can be used to compare funds.
Owen said: 'We are redefining our process here and the way we look at risk. In its crudest measure you can look at volatility and a lot of people do look at the volatility in the newspapers attached to every company, which is standard deviation over a specific period.
'It would certainly have told you this time last year that Invesco European was a highly volatile fund. In the UK sector you would see ABN Amro UK Growth, Solus UK Special Situations and Dresdner UK Mid Cap as highly volatile funds.'
That is borne out in volatility figures up to 1 May, according to Standard & Poor's, which show that the Invesco fund has a historic volatility of 10.04 compared to a sector average of 6.53. The Solus fund has a volatility of 9.02 and the Dresdner fund has a score of 9.27 compared to the sector average of 4.69.
But, volatility as measured by standard deviation provides only a partial and limited picture of the level of risk within a fund.
Using discrete periods, Owen said, can pick up any changes in the pattern of volatility and hence indicate whether a fund manager's style is consistent or whether there is an increase in the risks being taken within a fund. It has to be remembered that risk is only a snapshot. Low volatility can indicate a consistent approach, something best seen by comparing volatility over a discreet period. Over a single period however, low volatility can be unintentional.
Owen said Plan Invest was looking to move beyond using it as its sole risk measurement criteria. He said: 'What we are after is some sort of more complex system looking at beta and Sharpe ratios over discrete periods to see if there is any consistency in our measurement of risk.'
Moving into the use of new measurements has some barriers. As well as integrating them into a fund selection process, there is the issue of obtaining the information. Some groups like Threadneedle offer a lot of information openly to clients. Others like Fidelity are much less forthcoming, Owen said.
'Some of this information is quite hard to get hold of. We are looking at a number of factors that will allow us to put a Plan Invest risk rating on every fund after having looked at Sharpe Ratios and volatility over distinct periods,' he said.
One method that is sometimes used is to plot scattergraphs showing volatility against returns, a method which, if funds are compared over different time periods can show consistency of the relationship between returns and risk taken.
Another simple tool for indicating the risk of a fund is to compare the alpha and beta of funds in similar peer groups.
The alpha of a fund is the outperformance of a fund or a share compared to its peer group or benchmark. A negative alpha number indicates underperformance against a benchmark over the period covered. Like volatility, alpha and beta are backward looking, based on historical data.
The beta is a measure of the sensitivity of a share or mutual fund to a benchmark, for example the FTSE All-Share or the Micropal peer group. If a fund's beta is 1.5 then it suggests the fund will go up 1.5 times any rise in the benchmark index or go down 1.5 times any fall. For a fund with a 1.5 beta, a 10% rise in the benchmark should theoretically result in a 15% rise in the funds asset value.
Alpha and beta figures are generally available from asset managers, although some are more forthcoming than others, however for each of these you have to ask questions about whether the fund has a suitable peer group or benchmark.
Richard Boardman, director, risk modelling and performance measurement at RSA Investments, said: 'What the beta tells you is how volatile relative to the market, in general, the fund is. If it tends to be more volatile than the market, then the beta will be bigger than one. It is therefore saying if you are very bullish about the market in general you try and leverage to a greater extent and want something that is going to go up faster than the market.
'It gives you a good sense of risk. If your bet about the direction of the market is wrong and the market falls, it is likely that a fund with a higher beta is going to fall faster than the market itself. If you are a little bit more sceptical about what the market is going to do, it might be better to go for a fund with a beta that is closer to one.'
Comparing alpha over discrete periods gives a view on how consistent returns are. Similarly, comparing beta figures gives a view of consistency and fund manager style with higher beta indicating a growth style.
Like Owen, Boardman believes that intermediaries are going to have to show that they are adding greater value in the future than simply using volatility. It is something that he expects clients, who are growing in knowledge and needs, to demand more and more.
Boardman said: 'Most people's attitude to investment is to look at a newspaper, see the past performance and base a lot of their decisions on past performance. I think advisers, if they are to add value, have got to start saying hang on a minute, Invesco European is a very volatile fund, shouldn't I be recommending Investec European managed by Albert Morillo, which offers much lower volatility? I think advisers have got to get into that field.'
Going beyond these simple comparative measurement tools allows intermediaries to dig in greater depth and get a perspective of the manager's skill.
As time goes on there will be an increasing number of sources of comparative risk information.
Morningstar's free website at www.morningstareurope.com is seeking to give a low, medium and high risk rating on each fund depending on its underlying portfolio and also provide standard deviation and sharpe ratio. It, however, has not yet got complete information on funds and so the service still has many gaps.
The FT Group, as reported in Investment Week last month, is seeking to provide a freely-available risk-adjusted ratings service for UK Oeics and unit trusts via the internet offering comparative risk reward histories for funds and award risk ratings based on them.
Threadneedle uses tracking error as a way to distinguish different funds within its range. Malcolm Kemp, head of quantitative research at Threadneedle, said that tracking error provides a risk measurement tool, that can be calculated as a forward-looking estimate or a historical track record.
However, Kemp acknowledges that it isn't always a guide to future performance and gives indicated returns against a benchmark and is therefore a relative tool. Its European offerings, the European Growth and more aggressive European Select Growth funds, offer different tracking errors, allowing investors to gear into the market with different levels of risk.
Kemp said this means that in times when the market is rising, the European Select Growth fund should outperform the European fund but in a falling market it will fall faster. It allows decisions to be taken against the market with bearish investors possibly more suited to lower tracking errors and investors with high risk tolerances better suited to higher tracking errors indicating more aggressive funds.
Boardman said: 'We can look at tracking error through the rearview mirror or the front windscreen. Looking through the rearview mirror, it tells you what the historical spread of returns is above and below the benchmark over a given period.
'It can also be forward forecast to show what kind of range of returns can be reasonably expected over a coming period. A tracking error of 2% would indicate plus or minus 2% on top or below the benchmark over a period of time.
'It shows how aggressively managed the fund is compared to the benchmark. The more aggressively managed funds will tend to have higher tracking errors which means that an investor going into these funds has to bear in mind the fact that it could be either good or bad but it is more likely to be a big number and big difference to the benchmark.'
Looking at a combination of risk and return in a more complex way than simple scattergraphs, there are tools such as the Sharpe ratio and Sortino ratio, which offer valuable assistance.
Kemp said: 'There are two ways of going forward: one is looking solely at the risk, the other is looking at a combination of risk and return. You can look at those in terms of relative returns or absolute returns. In terms of relative returns there is the information ratio.'
The information ratio is the ratio of outperformance against either the benchmark or peer group to tracking error. It offers a measure of relative returns achieved for the level of risk taken.
Kemp said: 'The information ratio is in essence telling you how skilled the fund manager is. If you asked the question 'Does Threadneedle exhibit skill?' then the way that I would do it is by looking at something that is essentially the same as the information ratio.'
Boardman said: 'Essentially, you want a large top number but with a low tracking error if you can. The higher the information ratio, the more we tend to associate that with fund manager skill.'
Over time, he said, the comparison of discrete periods shows a pattern of how that skill is implemented.
Sharpe ratios look at the return for each unit of risk taken. It is calculated by subtracting the return from the risk-free return on cash from the return of the fund. This number is divided by the fund's standard deviation. Therefore a fund returning 10% against a return on cash of 5% with a standard deviation of 10% will have a Sharpe ratio of 0.5.
Boardman said: 'If it is positive you have outperformed cash, if its negative you have underperformed, so for example last year the FTSE All-Share would have underperformed cash and therefore had a negative Sharpe.
Sharpes are basically looking for high returns going up in a straight line. A fund has a lower Sharpe ratio if it has higher volatility.
A variation on the Sharpe is the Sortino, which is a measure of downside risk. It is the Sharpe ratio taking into account only the downside volatility.
The top half of the equation is the same, but the bottom half of the equation ignores the upside, giving a comparative measure of downside historical risk against peers
There is a second variation on the Sharpe ratio that is often used, though far less easily obtained. It is the Treynor ratio, which is equivalent to the Sharpe ratio, with standard deviation replaced with beta.
Kemp said: 'With every measure you are always going to be able to pick holes in it in one way of other. No measure is going to provide you with all the answers, that measure simply does not exist. That's why we are always very, very careful to take a number of approaches when we look at the funds for internal risk management processes. Equally advisers would like to have more of these numbers at their disposal.'
As well as delivering an additional level of value to the client, including comparative risk measures in the fund selection process is a proactive move that could help intermediaries cope in the future, particularly, as Owen suspects, when that future could involve advisers constructing client portfolios against benchmarks.
Three years at Wells Fargo
Effective from 9 December 2019
One firm with permission suspensions left
Continuing the Architas education series for clients.
Needs to apply for authorisation