Studies have found funds can be too big or too small to outperform, suggesting size does matter
The assessment of whether to invest in a certain fund is a tricky business. Bearing in mind roughly 90 per cent of active funds do not beat their benchmarks over periods longer than 10 years, the odds are stacked against the investor.
So how can advisers improve
One area to look at is fund size. Generally, the theory is that once a fund gets too big it is harder for the fund manager to find investments that can accept these larger inflows. Some of the underperformance in the larger absolute return funds has been attributed to them becoming too big.
Similarly, there is the reverse situation where a fund is too small to pay for its fixed costs, such as personnel, accounting, custodians and so on. Funds such as these are often closed by providers or merged with those with similar objectives.
However, there are other factors at play. Alluded to by the FCA in its Asset Management Market Study, as funds grow and achieve economies of scale there has been a tendency for providers to not pass these cost savings to investors. Instead, active fund fees tend to cluster around key points and, compared with their passive counterparts, have remained relatively stable even post RDR.
With this in mind, it could also be that large funds might underperform if they are not reducing their fees in accordance with their size, which could be a more significant effect than issues relating to fund size and liquidity.
As a side note, the structure of fund fees is now being questioned, spurred on by the extra transparency brought in by Mifid II.
There is an argument that the current favoured model of charging regular ongoing charges figures leads to providers becoming “asset gatherers” aiming to achieve the largest fund size, instead of targeting outperformance.
Recently, a different model has been suggested using a fulcrum fee, where the investor pays a lower fee should the fund underperform compared with its benchmark. This better aligns investor and fund manager interests, although perhaps at a cost to simplicity or transparency.
Given the complexity of isolating and analysing a fund’s size and its impact on future performance, the next problem is gathering reliable data.
This problem is complicated by trying to compare like-with-like: the level at which fund size affects performance could be different for a small-cap versus a large-cap global equity fund, for example.
The investing style of the fund can also make a difference. A fund with high turnover will be more adversely affected by liquidity issues than a long-term buy and hold fund. Unfortunately, fund reporting standards are lower in the UK versus the US, hence most studies conducted on this are based on the latter market.
Studies of the US market indicate a mixed picture, depending on the construction of the study.
On the one hand, one study (Chen, Hong, Huang, Kubik, December 2004, “Does Fund Size Erode Mutual Fund Performance? The Role of Liquidity and Organization”, American Economic Review , vol 94) found performance is negatively affected as fund size increases, but can be counterbalanced by the overall funds under management of a firm increasing. The authors hypothesise there is an economy of scale within the firm in terms of shared security dealing and research.
Meanwhile, another study (from CFA Digest November 2012: Elton, Gruber, Blake, “Does Mutual Fund Size Matter? The Relationship between Size and Performance”, Review of Asset Pricing Studies Vol 2, June 2012) uses a factor model approach to show that larger funds tend to have higher alpha after taking into account the traditional Fama-French factors of the market index, stock valuation and stockmarket capitalisation. The devil is in the detail of how these studies have been constructed.
We have also undertaken some research, albeit with focus on multi-asset funds. Again, the difficulty here is that Mixed Investment and Absolute Return are relatively recent classifications.
In addition, not all funds report their data in as timely a manner as required in the US, so we were restricted to using proprietary-audited Defaqto data to perform our analysis.
That said, when considering risk-adjusted returns we found a weak negative correlation between fund size and subsequent performance over a two- and three-year period. Investigating further using other statistical tests, it appears fund size less than £25m and above £400m also indicates a negative impact on performance.
We recognise that more study is needed on this area, in particular for more granular data using monthly fund flows over longer time periods and considering other fund sectors.
Jason Baran is insight analyst at Defaqto