Are clients getting value for money from investment products?

Many investors are paying active fees and ending up with index-like returns, writes Jonathan Ramsay.

Whether clients typically get value for money from investment products has been increasingly questioned by financial advisers, regulators and the financial press alike, and is also very much on our minds at van Eyk. 

Delving into this topic does require some relatively technical analysis and commentary. Yet it touches on issues that advisers need to think about, in the same way that a football coach might need to work out the statistical probability of breaking the line with a particular combination of players. 

For instance, tracking error may sound like a bit of investment techno-babble, but if you are investing in a manager that you want to give you a return similar to – but a bit better than – the market, it is an important concept. 

VIEW ALL

View all

Tracking error measures the variance of investment returns from the return of the market, expressed in terms of the standard deviation.

For those who have forgotten about variance statistics or were never particularly interested, a high number, say 5 per cent per annum, means that the manager’s performance has typically been about 5 per cent above or below the market two thirds of the time. 

Chart 1 shows the tracking error of a fairly “classic” combination of three Australian equity managers – value, growth and core.

These funds were chosen in this example primarily because of their long track records, but this would be fairly typical of many model portfolios. 

Because ‘complementary’ investment styles have been used in this portfolio, you would expect it to have a lower tracking error and to look a bit more like the market.

You can immediately see that tracking error has fallen since 2000, both for the individual managers and the overall portfolio. 

There are two dynamics at play here – cross-sectional volatility and the size of fund manager positions.

The two periods when tracking error blows, the late 1990s and the GFC, are due to the former, when the level of ‘activeness’ of underlying managers might have been unchanged, but the dispersion and relative volatility of underlying stocks increased. 

Over time, however, cross-sectional volatility in the Australian market has not fallen dramatically. This supports the intuition that many managers have also become a bit more risk averse.  

One could argue that tracking error, or relative risk, is not in itself such a relevant outcome for your clients and that relative performance is more important.

However, there is no evidence that managers in general have become more efficient or are getting better at outperforming as a group. 

Technically, this relationship between the tendency to deviate from market returns and the ability to outperform the market is encapsulated in the information ratio - outperformance as a proportion of the amount of relative risk taken. 

In large cap developed markets, institutional investors regard an information ratio of 0.25-0.5 as a strong result. If our example retail portfolio was targeting a fairly modest 1.5 per cent added value per annum (after fees), then it implies a required information ratio of 1 (given a tracking error of around 1.5 per cent). 

This is something which the most successful hedge funds in the world would aspire to. To put this in context, the average Information ratio for the broad universe of Australian equity managers was slightly negative for the last three years, having been up around 0.5 at times in the last decade. 

At a more intuitive - and perhaps client-friendly – level, Chart 2 shows the extent to which the example portfolio’s current aggregate positioning differs from the market. The differences are very minor and so the outcomes are likely to be minor as well.   

One response is to up the ante and use fewer managers and/or products with higher tracking errors, so-called concentrated portfolios.

Chart 3 shows the tracking error of a portfolio consisting of three concentrated managers. 

In this case, the overall tracking error of 4-6 per cent per annum is more consistent with a goal of 2 per cent outperformance per annum and an information ratio of 0.3-0.5 (the average information ratio in the previous two years for the A-rated concentrated managers in van Eyk’s last review was 0.3). 

To an extent, this is going back to the future as a targeted tracking error of more than 3 per cent per annum would have been more common a decade or so ago amongst mainstream managers. 

However, it still points to the need to have more conviction in the managers that you use and hence more detailed research and a more granular understanding of the manager’s strategy.

In addition, one is picking managers from a more restricted universe and one inevitably ends up with some quite strong biases, which may not be the intention.   

Chart 4 shows the underlying positions of the concentrated portfolio versus the market, and immediately we can see that this is more punchy than the previous example as it has a very strong bias against large companies. This may or may not prove to be a good thing, but the bias was not intentional.

In fact it was a by-product of trying to pick one value, one style-neutral and one growth manager from the handful of A-rated, high-tracking-error managers available in Australia. 

The problem would be worse if constructing this portfolio from the funds available on a particular platform or approved product list.   

van Eyk believes there is a better way. It involves taking a view on which areas of the market you are more likely to find managers that can deliver a higher information ratio.

This may be because their style is more or less suited to the current environment or because that area of the market is less efficient. For instance, there is much more to manager style than value versus growth. 

A quality bias is an equally important way of tilting portfolios. We have had a bias towards quality (typified by companies with strong balance sheets and sustainable earnings) for a few years. 

We now believe that this has run its course and that there is better value in, for instance, turnaround stocks. This is precisely because the market’s focus has been on the search for quality to such a high degree – and now value is appearing in other areas of the market. 

We also believe that some of the high-turnover, quantitative strategies are ideally suited to the volatile, range-bound markets that we have seen recently and that we expect to persist.

Chart 5 shows the broader palette of strategies that we use to construct portfolios and which we believe gives us greater flexibility to: 

• Take a proactive stance towards market conditions by actively tilting portfolios (a Market-Driven strategy); 
• Bias the portfolio towards the type of specialist strategy or process that we believe has a higher probability of outperforming in the current environment (Specialist); and 
• Vary market exposure in order to allow for the client’s absolute risk tolerance (Absolute). 

This approach does require advisers (but mostly ourselves) to undertake greater analysis and then take a firm view, but we believe the benefits are threefold.  

If we can focus active risk in areas where managers are more likely to add value, we will increase portfolio efficiency. 

Secondly, this framework allows us to help calibrate portfolios to the preferences of the client by varying parameters such as overall fees or market sensitivity.  

Lastly, it leads to a more differentiated and better articulated portfolio position, which in turn helps advisers demonstrate their value-add and their edge. 

But perhaps the most compelling reason to look at portfolio construction from this perspective is that the vast majority of retail portfolios, because of their low tracking error, are currently constructed in a way that is destined for failure. 

Jonathan Ramsay is head of asset consulting research at van Eyk Research.