Peter Lynch, writing in the September issue of Worth magazine, makes a claim that cannot be supported by reality: that an investor can commit 100 percent to stocks, ignore the ups and downs of the market, and withdraw a constant amount of money equal to 7 percent of the original investment--- without losing principal.

Last week, we showed that when you tried that strategy with the Standard and Poor's 500 index, the Dow Jones Industrial Average, and three Lipper mutual fund indices, the investor would have lost principal HALF the time and would have lower results than what Mr. Lynch says is a "worst case" about two-thirds of the time. The results from last weeks study are shown below.

Crash Testing The 1995 Peter Lynch Strategy

Measure year High Median Low Loss Periods
S&P 500 1960 857,075 152,722 49,856 (6 of 16)
Dow Jones 1960 812,212 97,037 0 (9 of 16)
Lipper Growth 1965 737,297 19092 0 (9 of 11)
Lipper G&I 1965 829,215 195,228 53,074 (3 of 11)
Lipper Balanced 1965 484,215 106,225 0 (5 of 11)
Source: Towers Data Systems

The temptation to follow Mr. Lynchs' advice is buttressed by logic as well as Mr. Lynch's reputation as a master stock picker. Stocks DO provide higher returns than bonds. Bonds, particularly bond funds, have not met investor expectations. And it does seem logical that you SHOULD be able to withdraw 7 percent a year from an 11 percent return investment and not lose money.

The problem isn't logic. It's using "averages". Like the story of the man who drowned in a river with an average depth of one foot, stock investors can "drown" by investing in a time period where the total return from stocks is LESS than 11 percent. In fact, there have been 16 periods of 20 years between 1960 and 1994--- and in 10 of them, the average annual return was less than 11 percent. In two periods, the average annual return was LESS than 7 percent.

Note that even these figures are averages. In fact, there were individual years where the returns were negative and other years where the returns were very high. What counts is what ACTUALLY happens each year... NOT what happens on average.

Now lets look at what happened if you had begun the Lynch withdrawal plan in another period of great expectations for common stocks: 1966. Investing $100,000 in January, 1966 and taking $7,000 a year in monthly checks, you would have ended the first year with $83,162. This means your withdrawal rate for 1967 would have to be 8.4 percent ( $7,000/$83,162), putting an even greater burden on your portfolio in the second year.

In spite of that, the portfolio struggles though the late sixties and early seventies, ending 1973 with $92,752. Then it falls to $70,895 in 1973 so the total withdrawal rate for 1974 will be almost 10 percent. But the market falls still further so you end the year with only $46,078.

That means a withdrawal rate of 15.2 percent ( $7,000/$46,078)--- FAR more than stocks usually earn. The result of a few early bad years is that the portfolio is crippled: it ends 1994 at a value of $42,534.

A Close Look At A 7 Percent Withdrawal Program from the S&P 500

Year Start Value Withdrawal % Withdrawn End Value
1966 100000 7000 7.0% 83162
1967 83162 7000 8.4 95569
1968 95569 7000 7.3 98591
1969 98591 7000 7.1 83569
1970 83569 7000 8.4 78901
1971 78901 7000 8.9 82725
1972 82725 7000 8.5 90752
1973 90752 7000 7.7 70895
1974 70895 7000 9.9 46078
1975 46078 7000 15.2 55754
1976 55754 7000 12.6 61547
1977 61547 7000 11.4 50104
1978 50104 7000 14.0 46095
1979 46095 7000 15.2 47055
Source: Hypo/Towers Data Systems

Making regular withdrawals of principal is the equivalent of reverse dollar cost averaging: you are always selling MORE shares when prices are down than when they are up. The greater the gap between the dividend yield and the withdrawal rate, the greater the regular loss of principal.

Right now, for instance, stocks are providing an ALL TIME LOW YIELD of 2.4 percent. That's lower than the lowest yield in the roaring 20's. It is also lower than the 3 percent yields that prevailed in the late 60's.

In practical terms, this means a 7 percent withdrawal program is a commitment to selling off 4.6 percent of the portfolio, regardless of performance for the year. That would have been painful in 1994. No sweat in 1995.

Research by the Leuthold Group in Minneapolis shows that periods of low dividend yields are an indication of low future stock returns. When stock yields were below 3 percent, the average annual total return in the next ten years was only 4.9 percent; when stock yields were 6.1 percent or higher, the average annual total return in the next ten years was 14.5 percent, almost three times as much. These are AVERAGE results so it is still possible to have a good return starting from a low yield--- but the odds are working against you.

Skeptics should consider these results for the best and worst of the 10 year periods between 1960 and 1994: if you had started a 7 percent withdrawal program in the five years in which yields averaged 3.2 percent, your TOTAL average annual total return would have been only 3.4 percent, LESS than your 7 percent withdrawal rate. The same program begun during the highest yield periods would have produced an average annual total return of 16.5 percent.
Group Beginning Years Average Yield Average annual return
5 lowest yield periods 65,69,66,68,70 3.2% 3.4%
5 highest yield periods 82,80,79,83,78 5.1% 16.5%
Source: Towers Data Systems; The Leuthold Group

Does this mean you should "time" the market?

Mr. Lynch says no, and for good reason. Many try. Few succeed. The only thing you can do is be aware that you are playing against longer and longer odds.

For those who are accumulating, it doesn't matter much. But it you are retired and wanting to draw cash from your nest egg, you need to worry about the consequences of withdrawal programs when dividend yields are at historic lows.

Is there an alternative strategy?

Stocks and cash. Since 1926 there have only been 3 periods in which stock prices declined for more than a year. There have been 10 one year periods of decline. And 55 years of positive total returns. A money market fund "cushion" or a ladder of Treasury bills would allow investors to spend cash instead of stocks in a down year.

Next in Series: Finding A Safe Withdrawal Rate