Price-To-Earnings Ratio Backtest
For this next backtest, I decided to test one of the oldest value focused fundamental ratios in finance, the price-to-earnings (PE) ratio. The PE ratio measures the relative value of earnings, usually using earnings from over the past twelve months, to the current price of the stock. Extraordinary items are also typically excluded from earnings used for the PE ratio because by definition they are unsuitable and they are kind of rare at this point given accounting rules.
The price to earnings ratio for this backtest is calculated as follows:
PE = current share price / trailing twelve month earnings per share excluding extraordinary items
The disadvantage of this ratio is that it is not forward looking, but instead relies on the recent past. The advantage is that you do not have to forecast earnings or rely on analyst estimates. You can compare this retrospective valuation ratio to the forward PE ratio backtest from earlier in the week.
I decided to run a backtest of this ratio from January 2, 2000 to December 31, 2020 using annual rebalancing and dividing the universe of stocks into quintiles starting from highest PE ratio to lowest (cheapest) PE. I used my standard method of backtesting stock fundamentals on Portfolio123 that I described in an earlier article. The average excess returns for each quintile are displayed in the chart below. As expected, stocks with the highest PE ratio in the first 20th percentile underperform the S&P 500 equal weight index by 2.08%. Stocks above the 80th percentile (5th quintile) with the lowest forward PE produce an average annual excess return of 1.41%. There is a very linear pattern to average excess returns as PE decreases from the 1st quintile to the 5th quintile.
The full details of the backtest are in the table below.
Reviewing the average excess returns, we can see that this standard PE ratio underperform the forward PE ratio we tested earlier in the week. The 5th quintile average excess returns were a bit lower for the PE ratio and the 5th quintile average excess returns were not quite (-2.08%) as low as for the forward PE ratio (-3.17%). I did notice that the average excess returns for the 5th quintile in down markets is still positive (0.28%) in this backtest as compared to the negative return (-1.52%) for the forward PE backtest. I also noticed that the annualized return from January 2, 2000 to December 31, 2020 was only 10.73% for the 5th quintile, which is lower than the annualized returns for the 4th quintile, but only by a negligible amount.
It is interesting to note that the top quintile (lowest PE ratios) underperformed the bottom quintile over the past four years similar to the performance of the forward PE ratio. That seems consistent with the recent chatter on social media that value investing is dead. I’m not so sure I buy into that, but I just wanted to show you how that’s played out in the numbers.
Portfolio123 recently added several new valuation ratios. I’ll be backtesting several of those along with some old time standard valuation ratios over the next couple of weeks. Be sure to subscribe or follow along to get the latest updates.