It would be a strange situation if every time you wanted to pay for information, you had to buy and sell a load of shares. Yet we now find the more an asset manager turns over their portfolio, the bigger the slush fund they build up to spend on research.
I don’t like ‘going on about charges’ just for the sake of it. But it seems the charges story has a far more complex plot than many of us had thought.
Labour leader Ed Miliband’s intervention into the debate last summer brought the focus onto portfolio turnover rate. Yes, the Neptune fund quoted as the worst example of high portfolio turnover turned out to be a stellar performer. But Miliband was right to highlight the issue, which hints at the idea that someone in the City is making a lot of money out of portfolio turnover.
Of course, brokers make money out of their clients buying and selling shares, and like any other business, the more they do it, the more they make. And quite right too. But conflicts of interest are starting to emerge in the way brokers are paid in what looks like a less serious but still significant re-run of the soft commission furore of the last decade.
When asset managers buy and sell, they pay commission to the broker for the deal. This might be around 15bps, I am informed. But only about half of that cost goes to pay for execution. It now emerges that the other half goes to pay for research provided to the fund manager.
Nobody would suggest fund managers do not need research. But should that not be declared as part of the cost of managing the client’s money? The FSA currently thinks not. Its Conduct of Business rules say commission can be used to pay for research, provided it actually is research.
But those long-term critics of pension charges will not stop shouting about what they perceive as being a hidden asset management charge until it is reflected somewhere in factsheets. And who can blame them?
Yes it may be complex to show for a single fund, but a broad-brush picture of the level of these costs over the previous three years could be given.
And more fundamentally, why should half the commission a fund manager pays go towards paying for research? If it wants to buy research, why not go ahead and buy it?
An FSA paper published earlier this month highlighted this very point. That paper, Conflicts of interest between asset managers and their customers: identifying and mitigating the risks, spelled out how an investigation into asset managers’ practices around these commissions found most did not exercise the same standards of control over research bought through commission than they did when they paid for the research out of their own pockets.
Fund managers were not all found wanting. One firm is challenging brokers to justify why it should pay for other services. Another set a cap on research services spend, insisting brokers swithc to execution-only rates once this limit was reached. But many did not, simply directing commissions into a research slush fund that then goes to pay brokers for arranging all sorts of extras, many of which the FSA does not see as research.
Some of this cash was being used to give preferential access to IPOs or access to company management, both items that the regulator rightly points out are not ‘research’.
And while there is no evidence to date of any cosy agreements between brokers and providers over allocation of commission for research, as long as you have a pile of money waiting to be allocated, such suspicions are not going to go away.
The insurance and asset management trade bodies understand they need to get their members up to speed on the matter by the year-end.
It seems commissions are paid in this way because they always have been. I am yet to hear a good reason why research costs should not be unbundled from transaction costs.