Hi thread, someone I know wrote this today.
The S&P at 1000
With the S&P 500 Index back down to 1000 this week (September 20, 2001) - after dropping from 1300 last May and 1500 in early 2000 - it’s time to ask ourselves; “what can I expect to earn by holding stocks from this level?”
Rather than quoting historic returns, I think it’s more instructive to talk about the factors that underlie those return numbers. Let’s look at earnings, growth rates, profit margins, inflation, P/Es and interest rates.
Taking it all apart.
Earnings growth depends on growth in the top line and the profit margin earned on those revenues. Since we are talking about the aggregate of all US business, we can compare this to the growth in US GDP. This is what we find for the last 50 years:
S&P earnings have grown 6.2% a year.
Domestic after-tax profits have grown 6.8% a year.
Nominal GDP has grown 7% a year.
Inflation has averaged 4% a year.
This tells us that earnings have, for the most part, grown along with the American economy. This makes intuitive sense. It also tells us that profit margins have remained fairly constant over time if you average the good years and the bad. We’ve been through a decade of good earnings and earnings growth and look now to be entering a period of bad earnings. The question before investors is what is a normal earnings level at this point in time? Looking at the long-term channel of S&P earnings, I would conservatively say that a normal earnings level would be between $45 and $50 a share. There are some that would argue this is too low, but others who might argue that it is too high.
I think it’s reasonable to say that the next 10 years will bring us much of the same – *6% earnings growth. In the year 2010, that brings the earnings range between $76 and $85 for the S&P. My judgement is that nominal GDP growth won’t match the 7% figure because that figure included the inflationary 70’s, but nevertheless, a 6% growth of GDP and a 2 ½% inflation rate shouldn’t budge the earnings trendline from its customary 6% increase.
Putting it back together.
Now it gets a little tricky. What multiple do we assign to those earnings? An average P/E during those 50 years was 15.5. I think it would be a mistake to use that. The reason? Because, like I said earlier, the inflation rate going forward isn’t likely to be as high. High inflation reduces P/E rates via higher nominal interest rates. My feeling is that a 20 P/E is reasonable. This equates to a 5% earnings yield, which seems appropriate given a 2 ½% inflation rate. If you want to argue for higher inflation rates, then I would argue for higher earnings growth rates.
So, applying a 20 multiple on those S&P earnings gives us a level of 1520 to 1700. If dividends are 2% during that time, then total return from holding the S&P will be 6.8% to 8.1% a year. This may seem a disappointment compared to the 14% returns of the last 20 years, but that isn’t the right way to look at it. Instead, we should look at what else is available. Those who are fleeing to T-Bills can expect to match the rate of inflation. Those who will risk owning longer treasuries, will make 2%-3% more. If you go to riskier fixed-income alternatives, you may increase your return still more. If taxes are an issue, then the interest-bearing alternatives are taxed yearly as ordinary income, while the long-term gains on stocks have tax caps of 20% - with the possibility of dropping further. This makes stocks look pretty good for the rest of the decade. Those with long time horizons should look at things in this light.
In the short run, emotion and momentum will drive stock prices. In the long run, fundamentals, which are primarily earnings, will win out. Hope this helps with some perspective.
* Maybe higher, note the after-tax profit growth is higher than S&P growth as companies outside the S&P have grown earnings faster. |
