Ok, ‘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 could be improved. As I see it, there are three objectives when establishing an appropriate fee structure: attract and retain talented people; incentivise them to act in the interests of the fund investors; and split the spoils fairly between investors and the manager.
I’ve been mulling over previous comments and reckon this is the best fee structure I can come up with:
- A capped base fee equal to the minimum of: a) a fixed dollar amount, or b) a low percentage of assets under management. For example, for our fund the base fee could be the lower of $500,000 or 1% per annum. The idea here is that the manager gets enough income from the base fee to cover its expenses but doesn’t get a windfall gain simply by growing the size of the fund without performing. Obviously, while the fund is small, a fixed fee would be more than a reasonable percentage of funds under management, which is why you would limit it to, say, 1% per annum;
- A performance fee paid for outperformance of a benchmark, either a fixed hurdle or a relevant index depending on the type of fund. This means the manager only gets rewarded if they deliver performance over and above what the investor could achieve through a passive fund or bank account;
- If the fund underperforms its benchmark in any given period, that underperformance needs to be earned back before any performance fee gets paid. If the benchmark is a fixed rate, the hurdle should compound (i.e. you would need to earn back any underperformance from previous periods plus the benchmark return for the period in question);
- Any performance fee gets paid in units in the fund, aligning the interests of the manager with the investors in the fund. The manager cannot sell their ‘performance units’ inside 5 years but they do get to earn an investment return on the performance fees that have been locked up.
- Those performance units can be transferred back to existing investors in the fund if the fund underperforms the benchmark during that five year period. Obviously the investors in the fund when the fee was paid aren’t necessarily the investors in the fund when it’s given back, but the general principle is that, if you’re going to take some of the outperformance, you should give back some of any underperformance.
This is as close as I can get to meeting the three objectives listed above. Any suggestions for improvement?