Money managers under the microscope
Are marathon runners trying to sprint?
“The long is short. Investment choice, like other life choices, is being re-tuned to a shorter wave-length.” So stated Andy Haldane of the Bank of England in a speech last month.
If one of the key features of a mutual fund is that it is a long-term investment, then concerns that money is being managed over decreasing time horizons should be treated seriously.
This concern was made all the more potent as it followed soon after a European Commission green paper also pointed to this issue with this comment: “It appears that the way asset managers’ performance is evaluated and the incentive structure of fees and commissions encourage asset managers to seek short-term benefits.”
So what insights can be elicited from Lipper’s data?
Portfolio turnover of mutual funds in the UK provides insights both in the range of holding periods for different funds, as well as how typical holding periods have varied over time.
Among 1,142 actively managed equity funds for which annual accounts have been published and analysed in order to calculate portfolio turnover (the lesser of purchases and sales divided by a fund’s average assets over one year), the median turnover is 59.2 percent.
How widely portfolio turnover varies can be seen by analysing this data by decile, revealing that the least active, or most long-term funds, had a rate of 18 percent or less (the bottom decile), while the most actively managed funds achieved a rate of 170 percent or more (the top decile). Around 40 percent of funds (accounting for 483 funds) had a rate of 50 percent or less, while just over one quarter (nearly 300 funds) posted a turnover rate of 100 percent or more.
Putting this into the context of short- or long-term investments, a portfolio turnover level of 25 percent indicates a holding period of four years, 50 percent – two years, 100 percent – one year, and 200 percent indicates a holding period of six months.
So clearly there are fund managers that are changing their portfolios fairly dramatically. As a benchmark, it is useful to note that index tracking funds had a median portfolio turnover level of 11 percent.
Of course some of the higher numbers are likely to reflect activity where a fund manager has changed or even where an investment policy has changed – one-off events. In a future column we might look more closely at whether there is a correlation between these activity levels and the performance achieved over the long-term.
For now, what we do know is that greater trading activity will generate higher costs (even if these can be overcome by a fund manager’s performance). While trading-related costs are not disclosed as a single percentage, estimates from several academics — Wilcox (1993), Carhart (1997), James (2000) — suggest that likely costs will be 1 percent for 100 percent portfolio turnover. If stamp duty on UK equities is also included, the median cost estimate across all actively-managed equity funds in the UK comes to 70 basis points per annum.
One other point to note there is that while standard management fees and operating expenses tend to be much higher for retail than institutional fund investors, these trading costs hit all investors equally as they are borne by the portfolio rather than at a share class level.
What can also be shown is that despite an escalation in the concern being voiced, typical portfolio turnover levels have not varied dramatically over the past eight years that Lipper has undertaken this work, with turnover staying between 55 percent and 60 percent each year. Having said this, it does look as though trading activity slowed through the financial crisis but has picked up again more recently.
In other words, if there is pressure on fund managers to perform over short time periods, this does not look to have had a significant impact on average manager activity in the wildly different market conditions encountered in recent years.
But this is not to say that such pressures are not real. In separate analysis of cross-border European equity funds, my research suggests that for rolling periods of 1-year performance, there appears to be a correlation between sales and performance. This weakens somewhat for 3-year performance, while for 5-year performance the relationship virtually disappears altogether.
Such dynamics inevitably put pressure on fund managers, even if that pressure can be resisted by some, if not many, managers. This is not just pressure from investors directly, but also the knock-on effect on the business.
Marketing budget will very likely be weighted to funds with a good story to tell, be it good performance, a star manager, or a new fund launch. We should certainly admire those managers who refuse to chase short term performance by churning their portfolios even when such clear business pressures exist.
(Editing by Joel Dimmock) ((firstname.lastname@example.org; +44 20 7542 3218))