This section will go over the figures to show two things:
- the general cycle of stocks
- 10 year averages
As an investor you're a drop in the ocean compared to the stock market. Let's look at the stock market, 1900-1968:
Notable sections:
- 1900-1924, a series of 3-5 year runs overall averaging 3%
- 1929-1949, the "New Era" bull market, irregular fluctuations. Comparing the average level of 1949 to 1924, there's an annual growth of 1.5%, showing a lack of enthusiasm for stocks
- the bull run with small recessions till 1968, an average compounded rate of 11% no including dividends of around 3.5% per annum
Wall St went wild for this. Few people saw that the rise may have been overdone.
There was a decline from the 1968 to a 1970 low, around 30-40% (dependent on the index you look at), bu it was followed by a massive recovery and a new all time high.
Supplement this price data with earnings and dividends data.
There's a general picture of persistent growth. Only 2 of 9 decades show a decrease in earnings.
An investor can't really use this to understand the next 10 years of earnings, but it does encourage the need for a consistent policy of stock investment.
Look at the change in P/E ratios. In 1949 the S&P 500 was 6.3x earnings, in 1961 it was 22.9x.
Before looking at 1972, let's look at the previous stock market summaries. Starting with 1948's.
1948:
- not too high considering underlying values
- it's sitting at 275, gaining 50% over 5 years
- "from the standpoint of value indications"
- concerned that in 1953, the averages had advanced for a longer period than in most bull markets
- advice: cautious/ compromise policy
In reality, the market increased 100% over the next 5 years.
1959:
- at an all time high of 584
- "present level of stock prices is a dangerous one... prices far too high"
In reality, it fluctuated up and down. Growth stocks shrank, hot issues (IPOs of small enterprises) were pushed up by speculation.
1964 had 3 conclusions:
- "old standards (of valuation) appear inapplicable; new standards have not yet been tested by time"
- investor must base policy on major uncertainties
- "if the 1964 is not too high how could we say that any price level is too high?"
The principles of investment, the conditions of 1964 would call for the following policy (in order of urgency):
- No borrowing to buy or hold securities
- No increase in proportion of funds held in common stocks
- A reduction in common stock holdings where needed, to bring it down to a maximum of 50% of the total portfolio. The capital-gains tax must be paid with as good grace as possible, and the proceeds invested in first-quality bonds or held as a savings deposit.
Those investors following a dollar cost average plan and continue, or suspend until they feel the market is no longer dangerous. Don't initiate this plan at these levels, because many won't have the stamina to purse such a scheme if the results are unfavorable.
In reality, this caution was vindicated. The DJIA closed 1970 at a level lower than 6 tears before.
What can we learn from those efforts to evaluate stock market levels? Yes, pick a consistent, and controlled common-stock policy on the one hand, and discourage endeavors to beat the market.
“It is the mark of an educated mind to expect that amount of exactness which the nature of the particular subject admits. It is equally unreasonable to accept merely probable conclusions from a mathematician and to demand strict demonstration from an orator.” Aristotle
The Stock-Market Level in Early 1972
- the 3 year P/E ratio was lower than year end of 1963 and 1968
- now bonds yield twice as much as stocks, which offsets the better P/E ratio
- conclusion: unattractive for conservative investors
Perhaps there is a replay of the 1969-1970 decline, or another bull market fling, followed by a more catastrophic collapse
Commentary
Graham was prophetic
The heart of Graham's argument is that the intelligent investor must never forecast the future exclusively by extrapolating the past.
There was a lot of bull market baloney going around:
- stocks have been going up 7%, so they should continue
- stocks have always outperformed bonds, so screw bonds
- stock indexes ignore the companies that died. Survivorship bias!
- A study bu Elroy Dimson @ LBS estimate stocks are overstated by 2% per year, making it no better.
- if stocks are always going to go up, how you pay for them don't matter
If prices are higher, it's riskier.
In this period of baloney, ask skeptical questions:
- Why should future returns of stores always be the same as past returns?
- If every investor believes stocks will guarantee money in the long run, won't the market be overpriced? (And then how can returns be high?)
The value of any investment must always be a function of its price. The profits companies earn are finite, and show should your willingness to pay.
Focussing on rosy market returns will lead to "a quite illogical and dangerous conclusion that equally marvelous results could be expected fro common stocks in the future"
Everyone knows to buy low sell high, but will get this backwards. The more enthusiastic investors become about the stock market in the long run, the more certain they are to be proved wrong in the short run.
Some theory on the stock market's performance. There are three factors:
- real growth (earnings & dividends)
- inflationary growth
- speculative growth
In the long run, EPS has average 1.5% to 2%, inflation around 2.4%, dividend yield at 1.9%, meaning you could reasonably expect a 6% return.
To remove speculative growth, Shiller compares the S&P 500 vs average corporate profits (after inflation). When this ratio goes above 20, market delivers poor returns after; when it drops well below 10, stocks generally product handsome gains.
A second lesson on Graham's approach: the only thing you can be sure of (while forecasting future stock returns), is that you'll be surprised and wrong. Staying humble of your forecasting powers will keep you from risking too much. Lower your expectations but don't deflate your spirit.
In financial markets, the worse the future looks, the better it usually turns out to be.