Identifying skill from luck while selecting an investment manager
Identifying skill from luck while selecting an investment manager
About this video
Jason Hall from Hamilton 12 questions the widespread practice of selecting investment managers based on their past three-year performance. From his perspective, this methodology is similar to predicting the winner of a football tournament solely from three games. Jason views that rankings determined by a three-year track record don't necessarily give insights into future fund performances, as various factors can influence the outcome. He reveals that investment managers have an inherent preference towards value stocks or growth stocks which can lead to significant performance variations under distinct market circumstances.
To separate talent from chance, Jason advocates that investors must be pragmatic about expected profits, seek managers with substantial 'active share', comprehend the fee structure, and finally, evaluate the manager's consistent performance across diverse market environments. He wraps up by stating that being too fixated on fees can disregard some potential high-performing managers.
Find the full unedited transcript of this interview below:
0:00
Let's get in now to what separates a good investment manager from the rest. My next guest, who thinks that ranking them based on past performance is like tipping the winner of the AFL after three rounds of competition. Jason Hall from Hamilton12 joins me now to discuss what to factor in when selecting an investment manager. All right, Jason, I'm not very good at AFL in general anyway, but we should be ignoring that scenario or thesis, I'm guessing you're suggesting and what should we be looking at?
0:29
So picking an investment manager is one of the most challenging things that an investor can do.
0:36
If you look at past performance and you just rank investment funds according to a peer group and look at a three year track record, which is the most
0:47
weighted component of, say, a morningstar rating, ranking funds by their three year track record gives you almost no signal as to whether those funds are going to perform in the next three years. And as you mentioned, it's equivalent to, say, taking 24 rounds of a football competition, looking at three games and picking the top team is saying, well, that's the team that I'm going to back to win the Premiership. It is a very weak signal of ability. Lots of random things could have happened in those three games or three years for an investment manager. And researchers have been puzzled by this for a couple of decades and they tried to work out why is this short term measure of performance such a weak signal as to what we should do in the future? And the reason is, is that investment managers have a particular style that works for them.
1:38
But whether stocks of a particular style go in or out of favour
1:43
varies according to market conditions. So what do we mean by style? Generally, managers have a growth bias or a value bias in the way they put stocks into their portfolio. And if value stocks are doing really well, a value based manager can kind of ride the wave and earn higher returns relative to a peer group. And when growth stocks are doing really well, a growth based manager can ride the wave and do really well compared to a peer group. And so if you rank stocks by that three year performance, you end up picking managers who just by pure chance happen to be in the style of investment that did well during that three year period. So then the question becomes like, well, what can investors do to try to separate skill from luck? And there's really four things that investors can do.
2:29
The first is to sort of form a realistic view as to what it's likely to lead to be good returns over a long period of time. And if you look at the top quartile of managers of Australian equity funds over the last ten years, on average, that top quartile has earned 2% above a benchmark, say the ASX 200 or the ASX 300. So the first thing again, best you can do is kind to be realistic about what returns they can achieve. And earning 2% after fees is the realistic measure of top quartile performance. If a manager is promising 4% 5% outperformance, that can only happen with a series of very risky bets.
3:12
Okay. I'm just
3:14
going to ask you a little bit more about what you're saying with active shares and how outperformance can only result from a minimum active share. What do you mean by that? What sort of percentage are you looking at?
3:25
So what we mean by active share and style is if I'm an investment manager that is trying to beat a benchmark, some of my portfolio is going to overlap with that benchmark and some of it's going to deviate from the benchmark. And active managers in general have about a 30 to 70% active share. And what that means is that the other 30 or 70% of their portfolio is a direct replication of the ASX 200 or the ASX 300. Or in other words, the active share of the managers portfolio is where they're placing their bets. And so what they're trying to do is achieve their outperformance with those long and short positions effectively that deviate from the market index. Now that's really crucial in picking a manager because if you select a manager that doesn't have enough active share, it means it's nearly impossible for them to beat their benchmark. So for example, suppose a manager
4:24
says, we think we can beat the benchmark by 2%, but you observe that their active share, what deviates from the benchmark is only one third.
4:34
That means that to beat the benchmark by 2%, their bets, their buy and sell bets have to earn that a return of 6% in order to get that 2% average outperformance because the remaining two thirds of the portfolio is just a replication of the broader market. And what researchers have shown is that the only managers who successfully outperform over a consistent period of time are the managers who have a material amount of active share. So around about, you know, above the 50% threshold is what you should be looking for to work out. Is my manager likely to have any genuine prospect of outperforming after fees?
5:15
All right. You're looking as well in terms of how fees pay into how you might choose, I guess, an investment manager that's always going to be dependent on how much people want to spend, I guess. But what are you looking at in terms of the fee differential?
5:30
So that's also been a puzzle that researchers have looked into for a long period of time. So researchers have tried to work out how much fees matter, and this is a huge focus of the Government's superannuation reforms. The Government, with its my super benchmarking tests, is really trying to compress fees because the Government is very sceptical that active managers can outperform a benchmark and the majority of them over time fail to outperform. Once you take these into account. But some of them do.
6:00
And what research have shown is that fees are not a good signal of quality
6:05
in other things that we buy, whether we buy cars or we buy hotel rooms or we buy nice clothes. On average, the more expensive car or the more expensive hotel or the more expensive dress or suit is generally higher quality than the cheaper alternative. But that's not the case with investment managers. Our fees generally represent the cost of the inputs. So the cost of making the the suit as opposed to its its luxury value or the cost to making the car as opposed to its performance characteristics and fees don't correlate very well with performance. And in fact, on average, the higher fee managers after fees earn lower returns than the low fee managers. But that doesn't mean investors should have this very narrow focus on fees, because the difference between a top performing manager and a lower quartile manager in terms of fees is about 0.2% per year. Where is the performance differential between a top
7:05
quartile and a bottom? Quartile manager is about 3% per year. So if you rate managers by fees and say, I'm only going to analyse the managers who've got the lowest 10% of fees and really narrow your search down to those managers, you're necessarily removing from that search criteria some potentially very high performing managers. So the key point is that fees are irrelevant metric for investors and you want to avoid very high fee funds. If the fees are above 1.2% to 1.5%, it's going to be very challenging for the manager. But once you've analysed managers who have plausibly reasonable fees below that range of about 1.2%, after you get to that selection, it really has to come down to does my manager consistently outperform other managers with the same value or growth focus?
8:01
Does my manager have a consistent investment process and does my manager have sufficient active share to give that manager some reasonable prospect of continuing to outperform on the basis of stock selection? So most managers generate high returns from picking securities, that is, individual stocks as opposed to picking themes or industries or asset classes that are going to do well. All right, Jason, so you've touched on four things potential returns, persistence and style, active share and fees. I just want to get back to that persistence in style thing because like anything, if you persist at it for long enough, you would hope for some good returns.
8:38
That's right. And so what we what we want to see with the manager is, does this manager have an investment process which on average outperforms in different market conditions in rising or falling markets when growth or value stocks are going to outperform or underperform? Does the manager have is there evidence that the managers outperformance in one particular time period relative to the broader market is the result of identifying particular stocks that do well relative to other stocks with similar characteristics? So specifically, if you rank stocks, say by price earnings ratio and you just pick the stocks with the lowest ratios, that's a value bet. And some managers do that. And those managers are going to do well when value stocks in general outperform. But if they're going to persist with high performance over a long period of time, they have to identify the particular value based stocks relative to other value based stocks. And that's going to lead to a more consistent
9:38
outperformance and therefore better returns for the underlying investors relative to their benchmark.