An interesting article in FTfm, the Financial Times’ fund management industry section, caught my eye recently. The piece, based on new data from wealth managers SCM Private, argued that concentrated “best ideas” funds do not perform any better than more traditional diversified mutual funds.

The article concluded that “new data compiled by SCM Private […] suggest that, in the UK, at least, focused funds are no better than diversified ones”.

While the points were made well, it did not sway me from my preference for focused “best ideas” funds in order to profit from “true” return and higher Alpha generation.

During my Masters of Science in Risk and Asset Management, I read several academic research papers highlighting that diversification benefits are limited when the number of holdings in a portfolio increases to between 25 and 30 positions. Studies show that exceeding that ceiling of 30 holdings means that the portfolio becomes too diversified and can reduce the potential for alpha.

Therefore, concentrated “best ideas” funds allow investors to profit from increased alpha generation.

The FTfm article did prompt me to crunch some numbers, however.

SCM’s study defines a fund’s level of concentration as the proportion of assets its ten largest holdings account for. In addition to this definition, looking at the IA UK All Companies sector, and using the data gathered by FE, I also decided to use the number of holdings in the portfolio as a measure of fund’s level of concentration.

SCM found no meaningful correlation between a fund’s level of concentration and its annualised three year return. I decided to run the numbers and assessed if there is any correlation between a fund’s level of concentration and:

-          its active share (defined by 1-r squared)

-          its excess return

-          its Alpha generation

-          its information ratio

I also ran the numbers over a three year and five year period.

Let’s first highlight some of my results, using the number of holdings as a measure to describe the fund’s concentration. Using the number of holdings for 200 funds in the IA UK All Companies, it appears that a typical fund in the sector has 55 holdings. I defined concentrated “best ideas” funds as funds lying within the first decile of this population, in terms of number of holdings. The typical number of holdings for these “best ideas” funds is 28.5. Let’s now look at the results.

I also ran a regression analysis between the number of holdings and these four ratios. Similarly to SCM Private, I found no correlation between the fund’s level of concentration and excess return. I also found no meaningful relationship between the number of holdings and active share, alpha generation and information ratio.

I also ran the same test, using their definition of concentrated funds as the percentage of top ten holdings. Once again, I defined my concentrated “best ideas” fund as the fund within the first decile of the population sorted by the top ten concentration. The typical concentration for the top ten holdings in this subgroup is 51.50%. The average for funds within the IA UK All Companies sector is 36.99%. Please find below the results:

The regression analysis showed once again no significant relationship between the top ten concentration and these four ratios.

Reading these results, it appears that “best ideas” funds are actually more active than their peers. The managers run their portfolios with a higher company-specific risks and take less market risk. You are therefore investing in a “true” active fund. This active management style actually allows investors better access to alpha generation.

Nevertheless the increased alpha generation has not necessarily translated into a better returns – with returns in the “best ideas” funds not proving to be better than in traditional funds.

You will have to take into consideration the direction of the markets over the last three and five years in my analysis. The FTSE All-Share had an annualised return of 15.03% over the last three years and 11.01% over the last five years. Concentrated portfolios take on the bet of increased specific risk but with a lower market risk. The bet failed over the last few years as the direction of the market played out against “top ideas” funds. These funds were not rewarded by the market for taking active risk, as highlighted by the lower information ratio; a risk-adjusted return ratio.

Markets did not award stock pickers - so to put it simply, these results confirm the adage, do not fight the Fed!

This analysis was confirmed by comparing the betas of the “best ideas” funds against their sector peers. Concentrated “best ideas” funds actually exhibited a lower beta than a typical fund in the IA UK All companies sector, therefore highlighting lower market/systematic risk exposure. Over the last three years, the “best ideas” funds have had a beta of 0.87 against 0.91 for the sector. This trend is even enhanced if you are looking at the Bull Beta, i.e. the fund’s beta when markets are on the upside.

Nevertheless this exercise was also a good way to identify skilful managers.

Few managers actually succeeded in picking stocks that did better than the market, as their active risk approach was more than rewarded because their stock picking was highly successful.

We could use these ratios to screen funds for investors searching for highly risky funds with strong active bets. Three funds stand out over the last three and five years: GVO UK Focus, CF Lindsell Train UK Equity and Invesco Perpetual UK Aggressive. Please find below their performance:

This study confirmed the trend observed by SCM Private. In the UK All companies sector there is no significant relationship between the fund’s concentration and its annualised return. Managers running concentrated portfolios actually succeeded in generating higher active share and higher alpha generation, mainly because of an increased active/company specific risk. In an environment driven by liquidity injected from the central banks, this bet did not pay off – it was better to take market risk in order to generate higher return.

Charles Younes is an analyst at FE Research