The General Theory of Fee Relativity
The General Theory of Fee Relativity
Scientists often use every day analogies to explain complicated scientific concepts. For example, Einstein’s E=MC2 tells us that there’s enough energy in the mass of a sultana, assuming you could harness it, to power New York for a day. Or the fact that the nucleus of an atom represents more than 99.9% of its mass, but if you enlarged an atom to be the size of the Melbourne Cricket Ground, the nucleus would still only be the size of a pea.
These analogies are useful. But the best tip I received about trying to understand the Theory of Relativity or Quantum Mechanics is to accept that your brain is not built to comprehend them.
Just as your eye only responds to the small percentage of the electromagnetic spectrum that you need to navigate the world around you (the only difference between the visible light spectrum and microwaves, radio waves and x-rays is the wavelength), your brain only comprehends speeds and sizes that you need for everyday life. When you start talking about the speed of light, the size of the universe or the probabilistic behaviour of an electron cloud, these concepts are outside our brain’s spectrum. They’re no less real. It’s just that we don’t need them to navigate the world in which we live.
Which brings me to a problem I have with fee disclosure. ASIC has done a great job of standardising Product Disclosure Statements (PDSs) so that fund managers need to include a standard table of fees. This table must include examples and is supposed to make a comparison of funds a straightforward exercise.
But to the human brain, 1% and 2% are both minuscule numbers. Expressed as a percentage of assets, the fees seem small and the difference between two small numbers doesn’t seem particularly relevant.
Industry super funds have done a great job of showing how much difference these small numbers can make to your long term returns. But when you’re reading through a PDS and comparing it with a similar fund, most of us see 1% for one fund and 1.8% for another and think ‘well, the performance is going to be far more important than one or two percent’.
What I’d suggest is reframing those fees, not in terms of the total asset base but as a percentage of the expected return. Of course, expected return is a wobbly number, but ASIC could set some straightforward numbers to use as examples. Or it could make managers provide a range. So you would end up with a table which has fund returns from, say, 6% to 12% down one side, and the percentage of that return that goes to the manager down the other side.
Instead of forcing a fund manager charging 2% per annum to provide an example stating ‘if your average balance is $50,000 dollars, you’ll pay $1,000 per annum in fees’, ASIC could force them to say, ‘if this investment returns 10% per annum, the annual management fee will be 20% of your returns’.
When faced with the prospect of one manager taking 20% of their expected returns and the other 10%, retail investors might give the issue a lot more attention.
Custom Search 2
Comments
As I pointed out in a feature article many years ago (see below link), you can take it one step further and compare the fee to the excess return above and beyond the risk free rate. For example, and as you point out Steve, a 2% fee structure could be viewed as a 20% fee on the returns from a fund that averages a nominal return of 10%.
But let's say that as an alternative you can invest in treasury bills or secure bank deposits earning 5%, then the fund manager is actually taking 40% of your expected excess return for taking on the risk on investing in equities. Not a bad cut for the fund.
http://www.intelligentinvestor.com.au/articles/124/How-your-mis-managed-...
Hi Steve - I just got an interesting question from my father (both of us have units in your fund). He asked, hypothetically speaking, what is stopping you (or any fund manager) from walking away with funds under management? I thought that would be impossible however after reading about Trio Capital and their money vanishing act I wasn't quite as sure.
Cheers!
Couldn't agree more Steve (and Gareth), its a point often overlooked by 'investors'. Even when it is laid out infront of them and explained what impact this has to their wealth over a 10, 20 and 30 year period people still make decisions based on short term numbers.
On a related matter, I did wonder why II opted for an unlisted open-ended fund as opposed to launching a LIC, Although I assume this was largely an issue relating to FUM size
This may be a dumb question but why does the management fee have to be based on the invested balance?
Aren't there any other options?
Aren't there economies of scale?
For my SMSF, the management fees are 0.48% of my balance and that percentage will reduce as my fund gets bigger. (That doesn't include brokerage, just audit fees, tax levy of $150 and subscription to TII).
There are alternatives and, in the US at least, they seem to be garnering some support from those responsible for allocating cash.
I had a US fund of funds manager in here last week talking about a model where the manager gets a fixed based fee plus a performance fee. Here's $1m, enough to cover all of the operating costs, and you only get more if you deliver the goods. Perhaps that encourages too much risk taking but you're right - the investor should get some benefit from economies of scale. Especially given bigger is usually worse for the investor.
There are issues with unit trusts. From my perspective the biggest is that you are likely to get maximum redemptions when you least want them.
But they pale in comparison to the weaknesses of an LIC model. This area has been a disaster. The fixed costs of having a board and all the listing costs of an LIC make it almost impossible to generate above market returns for anything less than $30m. Plus you have the issue of them trading at substantial discounts to NTA. At least with unit trust you can always get your money back (or invest more) at the underlying asset value.
Argo and AFIC work really well, but it just doesn't make sense unless you're big.
Thanks for the vote of confidence Rowan. Pass on my thanks to your father for putting that thought in a public forum!
On a serious note, I would have said to you three months ago that the regulatory system in Australia makes this impossible. The custodian (NAB in our case) holds all of the assets, all of the time and all I can do is allocate the funds amongst different assets (ASX listed or soon-to-be-listed shares and cash in our case).
If ASIC called me and asked for proof of the fund's holdings, I could provide it in about 20 seconds and it could be audited back to the individual companies' share registers within a week (the accounts get formally audited once a year).
My limited understanding of the Trio situation is that there is a custodian, of sorts, Hong Kong-based Global Consultants and Services Limited. But the assets the custodian thinks it holds are proving very difficult to find.
Perhaps if you notice us change auditors to Lichter, Yu & Associates of San Francisco (Trio's auditor), it's time to start worrying.
As you point out Steve, each of the different fee structures available is a two-edged sword. For example, whilst performance fees should help to align the interests between managers and investors by only rewarding managers for good performance it can also lead the manager to take big bets, particuarly after a period of sustained underperformance to 'catch up'. This may result in an all-or-nothing or shoot-the-lights out approach with very nasty consequences if things don't go right (which coincidently probably won't given the managers previous track record which got them into that situation in the first place!).
That is a very enlightening argument Steve, showing the fee as a percentage of the return. It certainly highlighted to me the difference in fees much more clearly. I think the fund managers would probably object, as seeing figures like 20% might scare off investors, but I like the idea.
I think the model of a fixed fee plus a performance fee has some merit. I can easily invest directly in stocks such as STW (ASX200 ETF) to get the index, but I would be willing to pay high performance fees for a fund manager if he/she beat the index.
Fixed fee plus a performance component is relatively common. But as has already been pointed out - this can create a situation of excessive risk taking.
It's almost like a free option on the fund's performance, which obviously only pays out when a certain performance level is reached. As far as the manager is concerned there is limited downside (i.e. to the fixed base fee of say $1m) and unlimited upside.
The result? A willingness to take on excessive risk.
While this is an interesting and fair point you raise Steve about fee disclosure I can't help but notice such changes as you describe are as much in your interest as they are in the investing consumer.
"Reframing those fees, not in terms of the total asset base but as a percentage of the expected return" would be in your interest by making your fund appear more attractive by highlighting or even amplifying a cost advantage over alternative funds that use those excess fees to attract investors via payment to advisors or advertising. There is nothing wrong with self interest but it helps to know where someone is coming from.
Maybe disclosure should start at home – it is just an observation not so much a criticism.
Agreed but it could also be seen as a shameless spruik for Steve's fund if this had been raised upfront without Justin's note so... I can sort of empathises with the sensitivities around this particular point.
That being said, I think Steve is perfectly within rights now to not only disclose his fee structure but to also highlight his reasons for doing so.
Fees based as % of FUM presents the same moral hazards as executives on fixed remuneration plus performance bonus components. The manager is essentially getting a free option with limited downside and unlimited upside. What is to stop excessive risk taking even with a fixed % to FUM? To align manager interest with investors, I suggest the manager be paid a % of any performance over the risk-free rate, and there must be a high-water mark hurdle. In other words, if the risk free rate is 5% and you achieve only 1% this year, you dont get paid this year. And the year after, you only get paid if you achieve more than 9% (4% back to high-water mark and another 5% to match risk free rate).
So, the manager will say 1 year is too short a time horizon. Okay, my answer is set your time horizon ie 5 years, and you only get paid at the end of your time horizon if time horizon returns exceeds the risk-free rate over the same time horizon.
Interested to hear what others think of this.
Thanks for the answer Steve, I'll pass it on.
In the mean time I just told my old man I met you in Adelaide and you didn't seem like a crook, which was fine by him!
Anyway I think its great that you can publish & answer questions like this - I'm sure other people will agree it actually demonstrates more integrity.
The high-water mark method does have some good points, the only criticism I've seen is that if the fund is performing badly, the manager could close the fund and start again elsewhere as he is unlikely to get performance fees from the current fund. Of course, existing investors aware of this history would be unlikely to invest again with him.
I don't know if this is applicable to funds, but for direct shares I like the general concept of "skin in the game". When the CEO and directors own shares, you know that they are participating in both the upside and the downside.
I met an investment manager once who I liked and considered to be an astute and active investor. In drawing up a proposal I specifically said I wouldn't come into a flat percentage fee on funds invested as I had been burnt. I would be prepared to pay 10 or 20 percent or better of what he made for me. He came back with a flat percentage on invested funds and we parted company. Seems this type of fee is deeply imbedded in the industry.
Managers which have a significant amount of their own wealth invested in their funds shows a genuine attempt to align their interests with other investors. This way they are exposed to the downside as well as the upside. Unfortunately these managers are far and few between
The fact remains that a fund which can achieve a 12 percent return and charges a 2 percent fee is better value than one which returns 10 percent and charges 1 percent. The question of course is how confident you are that the expensive fund will actually outperform the cheap one.
One particular way of reducing the potential for excessive risk taking due to performance based fees would be to make the performance fee based on average out-performance over a set period of time...say average outperformance of "x" per cent pa over previous 5 years. 5 years may be a long time but as an example it makes excessive risk taking less attractive when the out-performance needs to be sustainable.
[...] ‘perfect’ might be stretching the friendship a little, but my post on The General Theory of Fee Relativity generated a surprising number of comments, most of them excellent suggestions on how the status quo [...]
Post new comment