Giving the carrot some stick
Jeffrey S Molitor, who retires this year after more than a quarter century at giant fund manager Vanguard, has that old-school bow-tie gravitas of someone with no time for fads.
In 1998 he wrote a letter to the US Securities and Exchange Commission warning against changes to the Investment Advisers Act 1940 allowing mutual fund investment advisers to charge one-sided performance fees.
"There is no evidence that performance fees improve investment results," he wrote. "The benefits of performance fees accrue only to advisers, who can seek higher fees with limited downside risk."
You can see where he was coming from. Would a fund manager really work harder if they got a cherry on top of their management fee?
Fortunately, the fund management business can be analysed for reasonably objective answers to this question.
One study of the effects was published in 2002 by finance professors from New York and Fordham universities. It found that funds with incentive fees tended to perform slightly better - an outcome attributed partly to better stock selection and partly to lower overall fees.
There was also a tendency to take more risk, particularly after periods of underperformance.
This sounds significant, although there is a big difference with US funds that means the results do not apply to New Zealand. In the US, mutual funds can use incentive fees only if they also carry a penalty for underperformance - a structure known as a fulcrum fee.
It works by having a base fee that forms the axis for a bonus or a discount, depending on results.
Investors probably find this idea appealing, but it seems fund managers do not.
A 2011 paper on fees by managed fund research house Morningstar commented: "The reaction of US fund managers to more investor-friendly fee structures does not provide a strong endorsement of their confidence in their ability to add consistent value. The introduction of these regulations resulted in the number of mutual funds charging performance fees falling dramatically."
Ironically, one of the biggest users of performance fees in the US industry is now Vanguard, a specialist in low-cost funds with about US$2 trillion under management.
In New Zealand, however, retail investment funds are free to charge performance fees with no downside - and many do.
The idea is to encourage fund managers to do better than average, although in practice it's hard to be sure they work that way.
Investors in the Marlin Global listed fund run by Fisher Funds, for example, may have found it difficult to understand why a performance fee of $732,947 was payable by their fund for the year to June.
Marlin was floated at $1 a share in late 2007 with a mandate to invest in international equities.
Bad timing perhaps, but its share price has been below $1 most of the time since, including all of the last financial year. Last week, its shares were trading around 81c.
Taken at face value, it looks like Marlin investors are paying extra for inferior results.
This ought not to happen because Marlin's fee structure includes a "high-water mark" which should mean performance fees can be paid only if the fund is worth more than its previous peak.
There is also a hurdle requiring the fund to return more than 5 per cent above the change in the 90-day bank bill index during the year. The index models the return from investing in short-term bank rates, so Marlin's benchmark is known as an absolute return, as opposed to being measured relative to a share index.
You could argue this benchmark is easy to beat when sharemarkets are rising, and you'd be right. Perhaps in mitigation, Marlin has something akin to a fulcrum fee, in that its annual base fee of 1.25 per cent of asset value is cut if the fund returns less than bank bills.
So how can Fisher get a performance fee if the fund is worth less than it was to start with?
Taking its reported net asset value, Marlin reached a peak of 108c a share in 2010, earning Fisher a performance fee that year of $839,000. At last balance date Marlin's NAV was 91c, 3c up on a year earlier but still well under water.
The answer relates to Marlin recalculating its previous NAV to take account of capital and dividends paid out. The adjustment apparently takes its previous high water mark back to 81c.
There are two problems with this. One is the sheer opacity of the process. I doubt there are many investors who could work out what fee would be charged.
The other is that it could be wrong.
Fisher's management agreement says "the NAV . . . will be adjusted such that the manager's entitlement to a performance fee is not affected by such dividend distributions". It doesn't say how, which leaves the adjustment method open to interpretation.
Personally, I don't see why either.
Taking a simple measure of what a share was worth at the outset, versus its current asset value plus dividends paid over the years, I get a total return of 24 per cent.
Had Marlin exceeded the return from stuffing the money under the mattress, plus 5 per cent, let alone its own benchmark, it would have returned well over 30 per cent in the time.
To me, this makes Fisher's performance fee look like money for nothing, and underlines the difficulty of constructing a fee formula that aligns a manager's interests with investors'.
The details of fee structure lead to some intricate arguments, and there is no single best solution, but surely fund fees should be simple and clear, while not rewarding managers for poor returns.
Tim Hunter is the deputy editor of the Fairfax Business Bureau.
Sunday Star Times