Yesterday, I attended a seminar hosted by EDHEC-Risk Institute. The speech was given by Raman Uppal, Professor of Finance, EDHEC Business School. The topic was “Why Does Equal-Weighting Outperform Value- and Price-Weighting“.

In the seminar, Professor Raman Uppal presented his research on comparing the performance of equal weighted portfolios with value weighted and price weighted portfolios, using data set from S&P 500 stocks.

The contents were rather technical and academic, but I will highlight a few good points raised by him

Simple is good

There were many researches done trying to optimize portfolios, but most researches were done with relatively short sample period and limited samples. Single starting point of the performance measures also raise bias of the researches. Actual Performance of Optimal Portfolios is very Poor. Professor Raman suggested that a simple 1/N (equal weighted) with re-balancing approach would outperform, but with higher volatility.

However, the higher total return with higher volatility implies that the reward over risk is still better over the benchmark, i.e. high Alpha.

Rebalancing is the key

Professor Raman went on further to explain the differences in returns. He believes source of extra alpha arises from “Contrarian Re-balancing each month” to maintain equal weights, which exploits the reversal in stock prices.  The higher alpha does not depend on the choice of initial weights.

Transaction Costs

To put the model to work, transaction costs must be taken into consideration. Professor Raman assumed 50 bp for buying or selling the stocks in his research. Interestingly, some attendees (fund managers) expressed that their actual transaction costs were higher especially in volatile markets. This re-affirmed my belief that cost is a very important consideration in portfolio management to deliver returns to investors.