Is Price/Earnings indicator useful for creating a high profitable portofolio?
Today, we will continue historical tests of indicators and look closer at P/E ratio. It is an indicator of cheapness, but we would like to know, how effective it is in pointing out “pearls” and helping in creation profitable portfolios. We will compare P/E results with former tests of P/BV and P/S.
All methodology of testing will be the same as in the former experiments, so if you’re not familiar with them, please go back and check (First article in this category).
We divided all stocks in 10 portfolios:
- Extremely cheap stocks
- Very cheap stocks
- Cheap stocks
- Quite cheap stocks
- Slightly cheap stocks
- Slightly expensive stocks
- Quite expensive stocks
- Expensive stocks
- Very expensive stocks
- Extremely expensive stocks
The first group contains stocks with the lowest positive values od P/E, following with higher values. The last two groups include stocks with negative value (remember, that company with P/E= -100 is “cheaper” than one with P/E= -3).
These are the curves of capital for each such created folio:
1. Extremely cheap stocks
10. Extremely expensive stocks
At the first glance, like in former tests, all folios seem to earn money faster, than FTSE All Shares, but let’s remember, that for our results to be reliable, we should compare them with un-weighted index (explanation). The annual profits for each folio are shown on the pictures above. Folio consisted of all stocks has got annual return of 10,45%. We can see the comparison in the table below:
For better understanding of these results, let’s look on the chart with annual profit over base values:
We can see a distinct and stable trend (folios 1-8). The conclusion is that cheapest folios (1-4) earn more, than expensive ones, moreover they perform a lot better than the un-weighted folio of all stocks. A shift from better than market performance, to worst occurred in the group number 5 (slightly cheap stocks). It indicates, that we should avoid stocks with P/E value higher than 14. The results for 9-10 folios are not compatible with this rule. It can be coused by the fact, that a lot of really expensive stock, which should be included in the last two folios are in such bad financial situation, that they were thrown out of LSE. They are no longer activly traded. Other reasons will be discussed later in this article.
Out theory looks good, but let’s check few other criteria.
Firstly, we are looking for stocks that present not only better performance than market, but their behaviour should be stable too. On the chart below, we can check how long each folio was above the un-weighted index of all stocks.
The cheapest stocks spend more time over market, than expensive ones, but the best results are shifted to the third group (cheap stocks). Moreover, the trend is not so clear on this chart and differences between groups are smaller. Folio number 3 will spend 75% of time above the index of all stocks. Strange fact is that even the quite expensive folios (6-8) beat market more often than 50% of the time.
The next important thing is diversity between stocks. We would like to avoid the situation, when small percentage of superb companies create good results. Missing few buy signals can turn our strategy into disaster. To check this, we would like to know, what is the distribution of “pearls” (stocks which will earn +50% in next year) in each folio.
We can see that the level of P/E has an impact on the “pearl” distribution. The first three groups of cheap stocks contain more than 10% of superb stocks. Groups 4 to 9 have got less than 10% each. Once again we can see, that 10 group stands out from the expected results. It includes a fair share of “pearls” just like cheap stocks, even more than groups 2 and 3. The reason for this is the same as in the first situation. 10th group includes stock with negative earning and among them can be such great companies as e.g. “Twitter” (had got negative earnings for a very long time) or with some temporary problems. They are not bad stocks, but have probably a negative earnings for some period of time.
Now, let’s see how much bad performers are in each group.
Once again, we have very similar trend. First three groups comprise more bad stocks, than groups from 4 to 8. The reason for that can be a fact, that cheap stocks can point out a bad companies, which don’t have negative earnings yet, but no one believes in their good performance in the future. Two last groups, with the most expensive stocks have got lot of bad performers too. It is logical, because they include stocks with negative earnings, so companies in huge trouble.
Let’s combine those two chart in one showing the difference between pearls and poor performers distribution:
The best ratio can be found in first two groups. There is a lot higher likelihood to find a “pearl” among very cheap stocks. Groups 3 and 4 contain almost the same amount of good and bad stocks. The rest groups, include more bad than good stock and of course 9th and 10th group are distinguished due to reasons we talk about earlier.
Last but not least – riskiness. On the chart below, you can find out how badly folios performed during their worst year.
Values are compared to the loss of un-weighted index of all stocks in the same year. It is clear, that cheap stocks lose their value very rapidly during bad years. We can see nice trend from 1 to 8 group. Moreover, group 7 and 8 lost only around 2,5%, because of that, they are very good place to wait for the end of the bear market. As we expected groups 9-10, which contain stocks of many troubled companies, so they are not a good investment at any time, especially when the market is going down.
In comaprision to P/S and P/BV, analysis of P/E gives the most starightforward forward results. Unlike P/S, it shows that the cheapest stocks are the best ones. In comparision to P/BV, it shows that the most expensive ones are not as bad performers as we can think.
Today, we have seen quite compatible results. Cheap stocks earn more, than market and contain more “pearls”. On the other hand, a lot of bad stocks are among the same groups. Unfortunatelly, higher profits go together with higher riskness; the lower the P/E level, the worst performanece during bear market. It was pointed out, that during hard times, it is good to look for stocks, which have level of P/E from 20-40 (group 7 and 8). Very important thing to remember it that, the two most expensive groups behave “strange”, because they include stocks of troubled comapnies and prospective, future market leaders, which don’t generate positive earnings yet. Our conclusion is that, we should buy cheap stocks.
Taking into consideration a lot of regularity we’ve seen today, the P/E is a very good indicator and can be used alone for effective market analysis.