Manager research is a complex task. After all, how do you determine whether performance is the result of luck, skill, or even temporary market distortions?

Here at Bellmont, we have developed a rigorous framework for selecting and appointing fund managers, evaluating two primary criteria: whether the manager exhibits genuine skill and whether the manager charges a reasonable fee.

The working title for our Bang for Buck methodology is ‘Ben for Block’, named after its creators. Ben for Block is one way to help us dodge the bullet of hiring overpriced, underperforming or self-proclaimed active managers that are just expensive implementers of a simple rules-based strategy.

If you can replicate the performance of an investment strategy with some simple rules, why would you pay high fees?

Obviously, this assertion of ours is met with much criticism from those fund managers that love to portray themselves as unique and active so that they can command higher fees. Our answer to these one-trick ponies is:

“If it walks like a duck and it quacks like a duck, then it’s a duck”.

After all, if you can replicate the performance of an investment strategy with some simple rules, why would you pay high fees? Also, no matter how good a manager is, if the cost of their outperformance is too high, our clients’ fee budgets are likely better used elsewhere.

We totally acknowledge that to achieve statistically significant results, we may need decades of data. Unfortunately, the shelf-life of a Chief Investment Officer is only a few years, so we do not have that luxury. Furthermore, many of the managers available to us simply do not have a track record going back to the 20th century.

Here is some additional information about how we compile our manager database and determine which managers are worth their salt.

Judging Genuine Skill

Firstly, how do we judge whether a manager exhibits real skill? For those of you that know Michael Block, he has a joke that he tells where, in reply to the question “how is your wife?” he answers “relative to what?”.

Despite this poor attempt at humour, this parable illustrates that the only way to judge whether a manager adds value and whether that excess return is true skill (e.g. Jensen’s alpha) is to compare a manager’s performance to a suitable benchmark.

This is where most investors get cheated by managers, namely by managers deliberately choosing an easy-to-beat benchmark to make out that they are better than they really are. Often the benchmark is plain wrong, sometimes bordering upon misrepresentation.

We find that excess returns are often something quite different indeed from pure Jensen’s alpha.

Ascertaining the correct benchmark is the most nuanced and complex part of our analysis. This is mostly because fund managers are incentivised to portray as much of their excess return as unique skill that should be highly rewarded. We find that excess returns are often something quite different indeed from pure Jensen’s alpha.

Let us look at a few obvious examples where the proposed benchmark is clearly wrong – a bogus bogey if you will…

  • An Australian equity manager suggesting cash as its benchmark – clearly the wrong beta, especially as equities nearly always beats cash.
  • A global equity manager suggesting the MSCI ACWI as its benchmark, whilst never holding EM equities – prior to 2025 this had been a free-kick for over a decade as EM underperformed DM.
  • A global growth-style equity manager suggesting the MSCI World index as its benchmark – clearly the manager has a structural style bias and most likely a higher market beta than the broad index.
  • A global equity manager suggesting the MSCI DM index as its benchmark, whilst its stocks – although domiciled in developed markets – generate all revenue from emerging market economies.
  • An Australian small-cap equity manager suggesting the Small Ordinaries index as its benchmark, when the index construction is so flawed that even my grandma can beat it.A shout-out to Patrick Hodgens and our friends at Firetrail who acknowledge this and set their performance fee with a 2 per cent hurdle above the Small Ordinaries index.
  • A manager that uses gearing suggesting an ungeared benchmark.
  • A private equity manager suggesting a preferred return hurdle of 8 per cent on drawn capital; calculating returns on drawn capital is dubious at best, the benchmark is clearly not index-relative and is clearly set too low.A shout-out to our friends at Partners Group who use an MSCI ACWI-plus-margin based performance hurdle.
  • An Australian bond manager whose excess return above the Ausbond Composite index can be entirely explained by systematically adding some duration and lower-grade credit.

That being said, we have also found that this analysis can work the other way around and make some managers look better than their performance relative to their own stated benchmark might suggest. This tends to be those managers with real skill whose style is simply out of favour, as we compare them to a more realistic benchmark that better represents their investable universe. For example:

  • A global value-style manager suggesting MSCI World index as its benchmark (clearly the manager has a structural style bias and most likely a lower market beta than the broad index, coupled with historic underperformance of the value-style factor).
  • An Australian small-cap equities manager suggesting the Small Ordinaries index as its benchmark despite being ex-resources.A shout-out to David Keelan and our friends at Ellerston who have outperformed the (more appropriate) Small Industrials benchmark despite underperforming the Small Ordinaries index recently, primarily due to the recent gold run and its impact on small-cap returns.

Ben for Block – The Methodology

How does this all play out in practice? We ask each manager for their long-term performance and their suggested benchmark. We then use a variety of fundamental and statistical methods to ascertain whether the benchmark is appropriate, or whether there is a more optimal benchmark to judge the manager against. If we find that there is an ETF that exhibits similar characteristics to the manager, we can use this as a proxy measure of the manager’s beta.

For example:

  • A growth-style equity manager outperforms the MSCI World index by 5 per cent. The MSCI World Growth index has outperformed the MSCI World index by 3 per cent. Given we can implement a simple factor bet and achieve the MSCI World Growth index return through a ‘smart-beta’ ETF, we would suggest that a fairer (albeit approximate) attribution of the manager’s 5 per cent excess return is 3 per cent beta and 2 per cent alpha.
  • In this case we would most likely not pay expensive active management fees for this manager and prefer to utilise a cheap ETF if we wanted growth-style exposure in global equities.

We then measure the cost of active management by the percentage of alpha that our clients keep, net of fees. We target a maximum of 30 per cent ‘alpha capture’ – i.e. the manager keeps a maximum of 30 per cent, our clients receive a minimum of 70 per cent. We also account for the cost of the cheapest ‘replacement’ for the active manager – for example an index or smart-beta ETF – which makes the comparison like-for-like from an implementation perspective.

Let’s look at a worked example for Australian equities. The cheapest passive exposure to the Australian market is the Betashares A200 ETF at 4bps.  If an active Australian equities manager charges 55bps in management fees, the real cost of active management is 51bps (55bps minus 4bps). Under this example, we would expect the manager to generate a minimum of 170bps (51bps divided by 30 per cent) in gross alpha over the long-run. We see this as a fair proportional split of the alpha between the manager and our clients.

Unfortunately, we see managers failing this litmus test all too often. A manager who charges 100bps to add 150bps of gross alpha is incredibly expensive alpha capture (66 per cent) and ultimately a waste of our clients’ fee budget – we can get better Bang for Buck (or Ben for Block) elsewhere.

So, who loves this analysis and who hates it?

LOVE:

  • Our clients, because it helps us improve total returns net of fees.
  • Active fund managers with real skill in stock-selection.
  • Active fund managers operating in areas of the market with relatively expensive passive alternatives (i.e. Emerging Markets), as the relative cost of active management is lower.
  • Passive / smart-beta fund managers, because we highlight how they can improve the efficiency of fee budgets and ultimately net of fees returns for clients.

HATE:

  • Active fund managers that have outperformed the broad index through luck, style-bias, gearing or any other strategy implementation which leads us to determine their excess return is not alpha (we’re looking at you, growth-style equity managers that have outperformed the MSCI World index but underperformed the MSCI World Growth index and charged v fees for doing so).

Stay tuned for our next article on manager performance benchmarking entitled ‘Bulgarian Weightlifter Theory’.

This article was originally published at i3-invest.