Updates and supersedes November 23, 2003 post.
Since the stock market took a dip over the last couple trading days, I thought it might be worthwhile to update my research regarding stock market valuations. In short, the question is: is the stock market overvalued?
I will use a modified Warren Buffett approach to analyze the situation.
Warren Buffett is the second richest man in the US and by far the most
successful US investor. Using my modified approach, the stock market
appears too expensive by about 23%. This suggests that you probably
shouldn't invest in the overall stock market, and the correction underway could continue.
Keep in mind the definition of investing. An investor pays money now so that in the future he will be paid a greater amount. The difference between this "greater amount" and what he paid orginally is called interest or return on investment, which is expressed in annual terms as a percentage of how much he originally paid.
An investor can invest his money in a number of interest-bearing assets, the least risky of which is a long-term US government bond, which currently pays 4.66% per year. Presumably, he can rest assured that the US government will always pay the interest, so investing in US government bonds is said to be "riskless." In order for an investor to invest in a riskier asset, he requires a premium to be paid on this riskless rate. For stocks of large US corporations, the risk premium that was required by investors since 1926 has been 5.0% per year on average (2004 SBBI). Given this risk premium, an investor investing now would require a 9.66% return on his investment to invest in stocks for large US corporations (4.66% + 5.0%). If he doesn't think he will get this return, he shouldn't invest. The question then becomes whether or not large company stocks are likely to pay 9.66%, given the current price.
Finding out how much large company stocks are likely to pay an investor is easier than you might think, if you use a couple of solid assumptions. First, as a measure of the combined value of large company stocks, we will use the Standard & Poors 500 Index, which includes the largest 500 publicly-traded companies in the United States, and represents about 80% of the total market value of all US companies traded on US stock markets. The index ($spx) is at $1,142. Second, we know how much profit the companies in the S&P 500 are making. In 2002, core earnings were $55.00. Let's assume that all of this profit is paid back to the investor.
Third, we know with fairly good precision how much profit 2004 would have produced if it were an average year for profitability. Over the last 70 years, corporate activity as a percentage of our economy has stayed very stable at around 51% (it is now 53%). So we keep that variable constant. Further, over the last 50 years, after-tax corporate profit as a percentage of our economy as measured by GDP has ranged from 3% - 6.5%, and averages 4.9%. It very rarely deviated from this range in the post-war period, as you can see from the following plot for 1929 through 2002 (NIPA Table 1.14, Line 13 divided by NIPA Table 1.1.5, Line 1).
In 2004, after-tax corporate profit as a percentage of GDP was 6.14%. An average year would have yielded 4.9% in profit, or 23.8% less than it actually did. If the S&P 500 were adjusted downward to an average year in equal measure as the rest of corporate America, 2004 core S&P 500 earnings would have been $41.92.
Fourth, we know with fairly good precision how much this profit will grow over time. Since corporate profit as a percentage of our economy is range-bound and corporate activity as a percentage of our economy is stable, average corporate profit will track our economic growth, which is likely to be about 5% per year -- 2% inflation and 3% real growth.
Using these assumptions in a spreadsheet, we see that an investment in these stocks is likely to earn about 8.7% per year. That's great, right? Well, not exactly. As detailed above, an investor requires a 9.66% return on his investment in order to be adequately compensated for the risk that he undertakes in investing in large company stocks. If he were to invest in the S&P 500 at these levels, he would be undercompensated for the risks that he is undertaking. That's no good!
This suggests that the stock market is expensive, and may be due for a long-term correction. My calculations put a fair price of $930 on the S&P 500 index while, as stated above, it is selling for $1,142, or 23% higher.
You could also make a reasonable argument that economic growth will be 5.5% a year (or whatever) rather than 5% a year, but this still wouldn't make the current S&P 500 price reasonable.
Sources: Please see the underlying US corporate profits spreadsheet and S&P 500 cash flow spreadsheet. I hereby release both spreadsheets under the Creative Commons License -- Share and Share Alike.
The risk premium is calculated from Ibbotson Associates' 2004 Stocks, Bonds, Bills & Inflation (SBBI), page 33. This serial is only available in print. The geometric mean is used for both government bonds and large-company stocks. The 2004 SBBI includes information from 1926 through 2003.
Please note that I used the figure for corporate profits that includes the inventory valuation adjustment and capital consumption adjustment, whereas I did not include these items in the profits in my November, 2003 post. The analysis doesn't change, but the numbers are a little higher when we include these items.
Comments