Friday, April 13, 2001
From: Mark Vakkur, MD
Re: Historical errors in "Negative Energy", April 9, 2001, Alan Abelson column.I have been a loyal Barron's subscriber for years, but frankly am growing tired of Alan Abelson's Chicken Little comments. Had one followed his perennially bearish comments, an investor no doubt would have stayed on the sidelines for one of the greatest bull market advances in history. Sure, he can smile and say, "I told you so" because in a very concrete and limited sense, he was right - markets do decline. Surprise, surprise. However, Abelson has successfully called 10 out of the last 1 bear markets; an investor who followed his advice would have probably exited stocks around Dow 3,000. The market would have to lose another two-thirds of its value for an investor responding to Abelson's gloom to break even with a buy and hold stock market investor.
His latest column took the cake and prompted this retort. In it, Abelson cites the newletter-writer Alan Newman who noted "that the unfortunates who bought at the top in '29 had to wait until 1955 to get even. Those who bought at the highs in '66 didn't get above water until 1995. And the unsuspecting souls who got into the market in December of '72 waited a full decade for a good night's sleep."
This is simply untrue. Using the Cowles stock market database (a capitalization-weighted index going back to 1871) and the S&P 500 (after it was created), I created a dividend-reinvested total after-inflation return index. The actual facts are presented in the following table:
| | First | Wait (years): | First Subsequent Month to Be Above Water for Good:** | Wait (years): |
| September, 1929 | November, 1936 | 7.2 | January, 1945 | 15.3 |
| January, 1966 | July, 1967 | 1.5 | January, 1983 | 15.5 |
| April, 1972 | August, 1972 | .3 | April, 1983 | 11.0 |
* Defined as the first subsequent month to show an after-inflation profit.
** Defined as the first subsequent month to show an after-inflation profit that was not subsequently followed by an after-inflation loss.
The following became clear:
- An investor who made a lump sum investment in the market at the peak in 1929 (I used month-end September, 1929) would have shown a profit by November, 1936, when the deflation of the 1930s is is taken into account. Admittedly, 7 years is a long period of time to wait to break even, but it is nowhere near the 26-year stretch cited by Abelson. True, the stock investor would have subsequently experienced a real loss as the market sank again, but his return was positive for good by January, 1945, a full decade before the time cited by Abelson, and 15.3 years after the initial ill-fated investment. Of course, this is a lump sum investor - a dollar-cost averager through that horrible period would have done much better, buying prices at bargain basement levels in the 30s. The stock market investor would have beaten the pants off of a bondholder over this period. If he had held on, would have shown an 814% profit (440% if inflation is taken into account) by September, 1959 (30 years later). By 1955, the year Abelson claims the "poor unfortunates" in the article were supposed to have broken even, they would have actually shown a profit of 533% (309% after inflation). If that is unfortunate, I hope I can have such bad luck!
- A January, 1966, lump sum investor would have needed to wait only 18 months for a chance to get out even, and 15.5 years until he was profitable (after inflation) for good. Abelson was off by about 12 years. Where on earth Abelson gets 1995 as a breakeven point eludes me; perhaps he was looking at an inflation-adjusted (but not dividend-reinvested adjusted) Dow. An inflation-adjusted S&P 500 broke even by 1992, but looking at the index alone is like looking at a bond's market price while ignoring its coupon payment; the S&P 500 would be at about 10 times its current value if it reflected reinvestment of its 3.5% average dividend yield over that period. By year-end 1995, our "unfortunate" investor would have earned 1,912% on his initial 1966 investment (317% after inflation). This is far from breaking even!
- The only factually correct statement in the column was that a 1972 investor would have had to wait a decade to break even (11 years, to be exact, after inflation). However, he would have had an opportunity to exit with a small, after-inflation profit just a few months after the 1972 peak (although the index was lower, the reinvested dividends would have lifted his total return to above zero). But this story, of course, had a happy ending: if the 1972 investor had held on, he would have multiplied his wealth by a factor of almost 29 by April of 2001 (even after the market decline off the March, 2000, peak) in nominal terms, almost 7 times after inflation. When you consider the extraordinary hyperinflation of the 1970s, this was no mean feat.
Several morals emerge from this story:
- Even if you had the misfortune to invest a lump sum at the very peak in the stock market, you would have done very well over the long run. In the very worst case, you would have to wait 15 years to break even. So what? If someone can reliably tell me when these periods occur, then I will try to sidestep them, but until then I will hold the course. (I challenge Abelson to calculate how a bondholder would have done during the inflationary 1970s or the deflationary 1920s - in every rolling 30-year period except the one ending in 1861, stocks beat bonds).
- However, even the gloomy lump-sum-at-a-peak scenario is unrealistic since most of us are not investing a single lump sum, but are dollar cost averaging into and out of the market over our lifetimes.
- Even if March, 2000, proves to have been a market peak of historical proportions, if the data above are any guide, then in the worst case we have between 10 and 14 years to wait until a dollar invested at the very peak shows a gain. So what? The S&P 500 could lose half its current value before a dollar invested in 1995 would show a loss, and 69% of its value before a dollar invested in 1990 would show a loss - and that's AFTER inflation!
- At any rate, it's a little late to be talking about market peaks. The return on any money invested from this point going forward is mathematically guaranteed to be about 30% greater return than a dollar invested in March, 2000, since the market is currently about 25% lower. Sure, the market may head lower in the short term (and by short-term, I mean several years) but the long-term record is compelling. The fact that we know stocks have outperformed every other asset class over the long-term combined with the fact that they are currently on sale should make us celebrate, not despair, the current market decline. If it was rational to invest for the long-term a year ago, it's even more rational today.
- You should check all of your sources before passing them on as facts - you do your readers a tremendous disservice by distorting the historical record.
I would be happy to share my data with you, to review all calculations, and to expand this letter into a feature article. My sources were Robert J. Shiller, your magazine (for S&P 500 closing prices and dividends), and the Bureau of Labor Statistics (for the CPI).
No comments:
Post a Comment