Most investors in the market deposit at least part of their capital in mutual funds. In Israel, hundreds of funds manage hundreds of billions of shekels. The difficult task facing the investor is which of the hundreds of funds to choose. An informed choice of the trust fund is of enormous importance not only for the investors, it also affects the efficiency of the allocation of sources in the entire economy.
Fund performance is measured by the Sharpe index which examines the average return of the fund in relation to its risk. Since the returns relevant to the investor are the net returns, i.e. after subtracting the management fees, the accepted indices for rating funds are based on the Sharpe index of the net returns. These are the ratings that appear on most websites.
The famous Morningstar star index is also based on the Sharpe index. Since in real time no one knows the future returns, the rating of the funds is according to the Sharpe index based on past returns. The problem is that this ranking provides very little information about future performance, as you can see in the top graph. The graph shows the relationship between past performance (X axis) and future performance (Y axis), and it shows that there is almost no relationship between past performance and future performance (except for the worst funds, which tend to continue to be bad). This is a sad situation, and it means that we choose funds more or less “blindly”.
Can the situation be improved? Is there a metric that can better predict future performance? My colleague, Prof. Richard Rolle from Caltech University, and I believe so. In the article we recently published, we propose a new fund rating index, called the SAS index: Shrikage Adjusted Sharpe ratio. The new index is a generalization of the well-known Sharpe index, and it greatly improves the ability to predict future performance, as can be seen in the lower graph. The bottom graph shows that there is a high correlation between the new index (also based on past returns) and future performance.
The idea of the new index is simple: the historical gross returns and the management fees should be considered separately, and each of these two factors should be given a different weight. This is in contrast to the method accepted today, which is based on the net returns, meaning that it gives the same weight to the historical gross returns and the management fees.
A simple example may illustrate the idea. Suppose we are debating between two funds. Fund A has an average gross historical return of 12%, and it charges a management fee of 2%, and Fund B has an average gross historical return of 11%, and it charges a management fee of 1%.
For simplicity, let’s assume that the risk in both funds is the same. Which fund do we prefer? Since the historical net return is 10% in both funds, according to the currently accepted approach we should be indifferent between the funds. But this approach ignores an important fact: the historical gross return is only an estimate (usually very inaccurate) of the future gross return, while we know the management fee. Therefore, management fees should be given a higher weight than past returns.
If we take a case where the historical return has no ability to predict the future return, it is clear that the past returns have no meaning, and we should only choose according to the management fees. In the opposite extreme case, where the historical gross return is a perfect estimate of the future gross return, the same weight is given to the returns and the management fees, and we become indifferent between the funds. The realistic case is in the middle: past returns have some predictive power regarding future returns, so they are weighted in the index, but their weight is less than that of the known management fees. The optimal weight to be given to management fees relative to past returns depends on how informative past returns are about future returns.
The optimal weight is related to the statistical theory of “shrinkage”: the more “noisy” the historical gross return (constitutes a less good estimate of the future return), the more it is “shrunk”, which means giving it a lower weight in relation to the management fee. In our research we looked at which contraction gives the most informative rating about future performance. We found that the optimal contraction is the one that gives a weight of (approximately) 5 times to the management fee in relation to the average historical return. This means that to justify a 1% difference in management fees, a 5% difference in the average historical return is required.
In the example where the management fees of Fund A are 1% higher than those of Fund B, we would prefer Fund A over Fund B only if its historical gross return was 5% higher than that of Fund B, i.e. Only if its historical average gross return was over 16%.
Only 8% of funds “beat the market”
Most of the mutual funds yield inferior performance to those of the market portfolio (which is actually the TA-125 index in Israel or the S&P 500 index in the USA. According to the latest statistics, only 8% of active equity funds manage to “beat the market” (yield a high Sharpe index than that of the market portfolio).
Is it possible to identify based on past data the funds that will attract the market in the future? If you use the conventional Sharpe index, the answer is no—strategies that invest in funds with the highest historical Sharpe indices yield significantly lower future performance than the market portfolio. On the other hand, strategies that invest in the best funds according to the new SAS index yield future performance that is about 1% higher per year than that of the market portfolio (standardized for risk).
Implementation of the new index in the rating systems of the funds may lead to a more informed choice between them. Turning the spotlight on the importance of management fees may also lead to increased competition between funds.
The bottom line
The Sharp index does not make it possible to identify, based on past data, the funds that will beat the market in the future. Conversely, strategies that invest in the best funds according to the new SAS index yield higher future performance
Prof. Shiki Levy is the head of the specialization in finance at the School of Business Administration at the Hebrew University