Background
We are not big fans of using the Price-to-Earnings ratio. We saw how funds that use the market PE to time the market are no better than a buy-and-hold strategy (sometimes B&H performs better,) and we followed that up with how every single “ratio” has a caveat. And PE is the dumbest of them all.
However, if a large group of market participants pay attention to single flawed metric, then there should be a profitable arbitrage strategy that exploits that anomaly?
Exploiting PE obsession
Researchers in the US figured out a way to do just that.
Active investors with limited attention and capital constraints use fundamental metrics to screen and sort potential investments. Price-earnings (P/E) ratios are extremely popular, and are typically calculated using four trailing quarters of net income. Changes in the rankings of published P/E ratios may influence investor attention and subsequent excess returns. From 1974-2013, decile long-short portfolios formed on characteristics of P/E rankings which are rebalanced monthly earn value-weighted monthly excess returns of 101 basis points with annual Sharpe ratios of 0.79. Decile long-short portfolios which are rebalanced daily earn value-weighted daily excess returns of 16.99 basis points with annual Sharpe ratios of 2.91. Excess returns are robust to size, value, profitability, investment, price momentum, earnings momentum, short-term reversals, and relative volume. Changes to a stock’s P/E ranking predicts excess returns even when the stock’s P/E ratio itself does not change. The return premium cannot be explained by fundamental risk, clustering of attention at round number P/E ratios, or autocorrelation in the regressors.
We haven’t tested this for the Indian market yet. But this is just too cool not to share!
Paper: Rankings of published price-earnings ratios and investor attention