When the FCA released its final report into the asset management industry in June last year the point that really struck me was the comment on closet trackers.
‘We estimate that there is around £109bn in ‘active’ funds that closely mirror the market which are significantly more expensive than passive funds’.
Unfortunately that’s all we got, so I got touch with the FCA multiple times to understand just how they had arrived at this figure – yet I didn’t receive a reply. Time went by and I eventually lost interest.
Then over the weekend the Telegraph reported that asset managers have to pay out £34 million in compensation to investors who were paying active fees for passive products.
If they are purely tracking the market then this is entirely justifiable. But just how did they isolate these 64 trackers? In the absence of a detailed breakdown of a fund’s holdings (which asset managers are only required to provide twice a year in their annual and interim reports) the primary measure for determining how much a fund correlates to the movement in the underlying market is called the R-squared or the coefficient of determination.
Without boring you witless, the figure goes from zero to one, where zero means that 0% of the movement in the fund is determined by the underlying market and one means that 100% of the movement is. ETFs and tracker funds typically exhibit an R-Squared of 0.99 and above over a time frame of one year or more, so 99% of the gains or losses of the funds are explained by the movement in the market.
To give you some perspective, over the past five years the R-squared of the average actively managed fund in the Citywire - UK All Companies sector to its Citywire assigned benchmark is 0.67 with only four funds higher than 0.9 and the highest observed figure 0.94 (see graph below).*
While you could argue that those around 0.9 could be doing more, these are unlikely to have been among the 64 funds.
The other measure which is harder to determine is active share. The term has become a buzz word in the industry over the past couple of years, as the active industry seeks to differentiate itself from passive products. This is holdings based and looks at the percentage of a fund’s holdings that mimic the index. There has been raft of evidence to suggest that the higher a fund’s active share (or the less the fund resembles the index) the better its performance is.
For the FCA to demand fund firms reimburse investors, then there must have been very little room for doubt that these funds were in fact trackers. If that’s the case, a list of the offending portfolios would be most welcome, so that they can be removed from investor whitelists and our active database.
And what then of the ‘£109 billion’ they referred to in June?
At the time that would have represented 9.7% of the £1.19 trillion actively managed assets tracked by the Investment Association.
We have obtained the PDF sent to the Telegraph (quoted below) and they are suggesting that this figure is the amount of assets in ‘partly active funds’. Though again no clarity from the FCA on what criteria is used to determine this.
The final report stated there’s £109bn in partly active funds charging fully active fees. And some observers inferred that this all related to closet trackers. This isn’t quite correct. This figure referred to the amount invested in partly active funds that were significantly more expensive than traditional passive funds. These funds may well be adequately disclosing how they are investing people’s money. But they are expensive compared to similar products, reinforcing the central finding in our study; that price competition in asset management is weak in a number of areas.
Be careful what you wish for
The precedent that the FCA’s ruling has set, means that we may see active shares increasing throughout the industry, but a high active share doesn’t always yield positive results.
Last year Jayesh Manek the competition winner turned UK equity fund manager hung up his spurs, following a 25 year career that saw his fund lose 56.4% after fees, while the FTSE All Share TR climbed 231.4%.
Manek’s fund was distinct in that it regularly featured among the very best and more often than not the very worst of his peers and it was always one of the portfolios with the highest tracking errors – if not the highest. This is an example of portfolio with a high active share that didn’t work.
There are times when his style of high conviction investment in Japanese small cap value stocks generates astonishingly high levels of outperformance – but there are also some very lean periods for the veteran manager. We are not in one of those periods and he tops the Equity Japan peer group over the past three and five year time frames returning 149% and 276.6% respectively.
While these are only two examples the point I’m trying to make here is that once funds go out of their way to differentiate themselves from the market they are trying to beat, the scope for blow ups increases as does the potential outperformance.
Your portfolio doesn’t need to be massively different to the market to generate outperformance and we are not at the point where investors hold a core of passive products with a sprinkling of ultra-high conviction, high active share managers as satellites.
This ruling may herald the end of the closet tracker and I doubt there are many in the industry that will shed a tear for these funds, but allocators will need better tools to identify those funds trying to differentiate themselves.
*A fund must have been active for a minimum of five years. This is the average of the one year rolling R-squared measures for each fund over the past five years, where the window is moved one month at a time, meaning that each fund has 49 data points going into their average.
Frank Talbot is Citywire's head of investment research