A better way to rate funds

Using conventional and quite popular mutual fund rating tools might help you justify your choice (in case you are managing other people's money) but it isn't a winning strategy to say the least.

A recent study by Morningstar confirmed that using only quantitative tools, that are based on past performance, isn't the right strategy.

The same study showed that ratings based on qualitative analysis worked 70% of the time. However, doing qualitative analysis demands more man power to complete.

There are plenty of traditional fund rating firms: Morningstar, Lipper, Citywire (ranks managers) and many more. They all base their ranking mainly on past performance. Since they employ a relatively small number of analysts they find it impossible to review critical factors on a continuous basis, such as the incentives and motivation of the management team, capacity and limitation on size, active share, etc.

Furthermore, current fund ranking methodology has plenty of problems. Here are just a few examples:

• The past performance achieved by a certain Fund Manager is still taken into account after the Fund Manager has been replaced with a new one.
• A fund that invests in a market that is less liquid most likely will have a hard time achieving the same results once the fund grew in size. Most fund managers tend to declare a soft or even hard close of the fund in cases where they fear that the fund has reached its capacity, however, in more cases than none, its way to late.
• Leading the peer group in terms of past performance might be a result of an investment strategy that significantly deviated from the strategy taken by the rest of the peers. For instance, if a fund manager decides to hedge part of the currency exposure, the end result might be very positive, however it might not be as a result of pure stock or bond selection.

The factors above, and many others, can only be taken into account once a proper qualitative analysis is performed. Conducting such a deep and ongoing analysis can only be done by utilizing a large group of people.

Also, checking that what is written on the fact sheet fits the investment strategy the fund manager has taken in the past, or simply verifying that the same fund manager is currently holding the steering wheel and is not involved in managing other funds or staff that are not related to the fund, are tasks that cannot be achieved properly by using quantitative tools.

In recent years the idea of using crowd sourcing in order to improve service and product ratings has become main stream as companies like TripAdvisor, Yelp, IMDB and Zagat have become extremely popular. It has been proven time and again that using the "wisdom of the crowds" works.

My favorite example is the results collected from the game show 'Who wants to be a millionaire?'. During the game participants are offered to ask a friend (who they consider to be an expert) or ask the crowd of audience in the studio. The data shows that asking the expert results in 65% success rates while the crowd is right 91% of the time.

More recently this idea has entered the financial world with companies like eToro, Motley Fool, Estimize, Vetr. It is also worth mentioning two pioneers in this field, Marshall Wace and what is now called TIM group.

What all these Fintech startups have in common is the idea of using crowds and their knowledge in order to offer improved financial ratings and market predictions.

So, on the one hand we have a problem with how funds are currently rated and, on the other hand, we know that wisdom of the many is advantageous in any selection process it has been applied to.

The only way forward is to leverage the knowledge and experience of the global investment advisor and fund selector communities to create alpha.

Want to know more about creating alpha? Oren Kaplan will be presenting Searching for alpha with Big Data at FundForum NextGen Distribution taking place in Boston, 25 - 27 October 2016.

To learn more about crowd-sourcing alpha creation to help select winning funds, go to https://www.sharingalpha.com/.