Warren Buffett's fortune is enough to stupefy anyone. Starting from scratch, he has amassed a fortune of billions and billions of dollars. And the most amazing part of that feat is not even that - it is the fact that he has achieved so much wealth in his lifetime purely by investing in the stocks and bonds of companies.
As he has mentioned a number of times, he credits much of the framework with which he invests to Benjamin Graham, his mentor and teacher from whom he learned how to invest. Thus, it should be of immense interest to anyone who wants to invest wisely, to hear what Graham has to say on the subject of the maximum price one should pay for buying a stock. After all, this is a perennial question that comes to the mind of investors - What is the right price to pay?
For the purchase of a stock to be successful, every investor relies on future earnings of the company and not its past earnings. But at the same time, Graham was of the firm opinion that when evaluating a stock and its future earnings, one can be conservative only by basing this opinion on company's actual performance over a period of time in the past. Thus, in most cases, the investment (and not speculative) value of stock can be arrived at by taking into consideration the company's average earnings over a period of five to ten years.
The company's profit in the most recent year may be taken as the base for arriving at the value in some cases, but only if it meets the following criteria -
(1) general business conditions in that year were not exceptionally good
(2) the company has shown an upward trend of earnings for some years past
(3) the investor's study of the industry gives him confidence in its continued growth
And only in the extremely rare and exceptional case should one rely on the assumption of a company achieving higher earnings in the future while calculating the price one pays. Higher future earnings should be taken into consideration only if it is a 100% sure thing, which is very rarely the case.
The above was a discussion of Graham's suggestion of which earnings should one take as the base when valuing a company by using a P/E ratio. Now for the second part - what would be the right multiple one should give those earnings to arrive at the price one should pay for the stock?
A conservative investor may rightfully give a very attractive company a higher multiple. This may be a company whose latest earnings are above its past average, which has extremely promising future prospects, or has an inherently stable earnings power. However, at the very heart of Graham's argument was his opinion that there must always, in every case, be some upper limit of this multiple that is assigned to the stock in order to stay conservative in one's valuation. He suggested that about 20 times average earnings is the highest price that can be paid buying a stock from an investment perspective. While this is the maximum one should pay for a company considered to have very good prospects, about 12 or 12.5 times average earnings would be suitable for the typical company with average prospects. This is because investment, as opposed to speculation, necessarily requires demonstrated value, which can be verified only by way of average earning power in the past. A P/E multiple of 20 in effect means an earnings yield of 5% (1 divided by 20).
It would indeed be very hard to conservatively justify average earnings of less than 5% of the market price of a stock without betting your money on an increase in earnings of the company in the future. Thus, according to Graham, a price to earnings ratio of higher than 20 times average earnings cannot by any means provide the margin of safety that an investor should have. It might be accepted by an investor in expectation that future earnings will be larger than in the past. But such a basis of valuation would then have to be termed "speculative". That is because speculation derives its basis and justification from potential developments that differ from past performance.
By the above thumb rule of not paying more than 20 times average earnings, Graham did not imply that it would be mistake to do so. He suggested instead that such a price would be speculative. Further, it should also be noted that such a purchase can easily turn out to be highly profitable, but in that case it will have proved to be a merely fortunate speculation. And very few people are consistently fortunate in their speculation.
Hence people who habitually purchase shares at more than about 20 times their average earnings "are likely to lose considerable money in the long run" according to Graham. This is all the more likely because if such a mechanical check were not enforced, investors have the tendency to time and again give in to the temptation and lure of bull markets, which always find some or the other deceptively pleasing argument to justify paying extravagant prices for stocks.