Is DFM consistency a virtue or vice?

Graham Harrison, managing director of Asset Risk Consultants

Go to the profile of Independent Research
Mar 28, 2019
1
0

Many investment managers will tell clients they aim for consistency; if they can deliver “average” performance consistently, their performance over time will trend towards being first quartile. In other words, by avoiding mistakes, they will end up delivering excellence. 

That sounds great in theory but how do DFMs actually perform?

In this article, we examine two sources of return dispersion - that between DFMs and that within DFMs - and find evidence that the internal dispersion of private client outcomes within a single DFM is at least as high as the dispersion of returns between DFMs. Choosing the right DFM is only part of the answer.

Inter-manager dispersion

Each quarter, the 25th and 75th percentile returns for various time periods are published in the ARC PCI Report, available free of charge on www.suggestus.com. By examining previous reports, it is possible to build up a picture of the return dispersion between top and bottom quartile DFMs.  

For this article, we have looked at 12-month return dispersion for the ARC Sterling Steady Growth PCI universe. The following results are based on comparing the returns of the more-than 50 DFMs included in the index, calculated using over 50,000 underlying portfolios that have a volatility of between 60 and 80 per cent of world equities.

The white band on chart one plots the rolling annual range of returns delivered by DFMs which, over the past five years, have been as high as 15 per cent and as low as -5 per cent.

Now look at the red line that plots the width of the white band or the inter-manager dispersion. The difference between top and bottom quartile DFMs has remained consistent at circa 200 basis points per annum, irrespective of the absolute level of performance.

Sources of dispersion between DFMs are many and varied. As the PCI are constructed without using asset allocation constraints, they express a wide range of investment philosophies and approaches. It is therefore perhaps slightly surprising dispersion across them over the past five years has been so stable. However, it should be noted that, throughout this period, central banks around the world have sought to deliver monetary policies that have been supportive of financial assets.

Intra-manager dispersion

While performance averages for each DFM are a useful way of comparing, one of the main purposes of a discretionary portfolio is to allow the DFM to tailor that portfolio to meet the specific needs of each individual private client. As a consequence, it is possible for two clients of the same DFM to have significantly different outcomes, yet both achieve their objectives.

Using the data provided on each underlying portfolio, the internal dispersion can be calculated for each DFM. That has been done in chart two for all the larger DFMs contributing to the ARC PCI universe in 2018.

The average internal dispersion is around 200bps per annum, with around half the DFMs having a return dispersion in the range of 140 to 240bps.

That means, for a significant proportion of DFMs, their internal dispersion of returns is similar to or greater than the “average” return differential between DFMs.

The level of internal dispersion is a key metric for investors to consider when selecting a DFM as it provides quantitative insight into their investment implementation process. A key factor driving the level of internal dispersion is the extent to which a DFM is model driven or offers bespoke solutions.

For DFMs using a model portfolio approach, consistency of portfolio construction and implementation should lead to low levels of dispersion. By contrast, for highly bespoke firms, outcomes may vary significantly by region, office or individual manager.

Some level of internal dispersion is inevitable: constraints on mandates, timing of deposits and withdrawals, fee scales and tax considerations all contribute to differences in outcomes. It is up to each investor to determine whether such flexibility is desirable or just adds to uncertainty.

Is consistency enough?

The effect of the two types of dispersion means the question of whether investors should be satisfied or not with their portfolio performance needs to be considered in terms of how they have performed relative to other clients of the same DFM as well as how the DFM has performed against others.

Chart three shows how the two type of dispersion interact by examining the 12 months ended December 2018. It shows the proportion of each DFM’s clients in each return quartile.

The analysis shows that, even for top quartile DFMs, a significant proportion of their clients would have recorded a third or fourth quartile return, and the reverse is true for bottom quartile DFMs.

Conclusions

So, what does all of this mean for investors?

  • Given the dispersion of return outcomes within DFMs is at least as significant as dispersion across DFMs, seeking to ensure the proposed mandate is likely to mean the investor will get the best a firm has to offer is crucial.
  • Once a DFM is appointed, they should be able to explain to their client both how and why their performance has differed from that of other clients of the firm.
  • If consistency of outcome is desired, a multi-asset fund or model-based solution may be the optimal route. However, the enormous benefit of a discretionary portfolio is that a DFM can take account of each client’s circumstances and tailor the portfolio accordingly.

Consistency of investment process may be a virtue but, as Oscar Wilde suggested, consistency is the hallmark of the unimaginative when applied to the implementation of a discretionary portfolio.

 

Graham Harrison is managing director of Asset Risk Consultants

DFM Directory
Outsourcing Partners

Money Marketing’s DFM & Outsourced Fund Solution Centre

Browse our directory of Discretionary Fund Managers, to manage your client portfolios, and Fund Providers, to search for a suitable one-stop shop fund solution.
Go to the profile of Independent Research

Independent Research

Independent research from carefully selected third parties.

No comments yet.