In two previous articles [1] [2], I described how I have been employing the secular trend concept for asset allocation in my 401K. I described what allocations I had made in my own 401K since September 3, 1999 when I moved completely out of equity in anticipation of the beginning of the secular bear market, which I believed had already begun the previous July. My strategy was simple, I hold only two assets, "equity" and "cash". I increased my cash allocation as the stock market moves "down" and increase it as the market moves up, with a deadband in the middle in which I do nothing. The equity component was the mutual fund American Century Ultra (TWCUX) which functions sort of as a cross between an S&P500 and NASDAQ index fund. The "cash" component is an asset-backed security fund that has been paying about a 6% return over the last 4 years or so and is still yielding (as of the last three months) about 5.5%. Like a money market, its share price is always a constant $1.

Since my last update in April my company's 401K has dropped TWCUX as a fund option and I have been forced to move my assets into an S&P500 index fund. The specific fund is an institutional fund which has no symbol, so I will use the Vanguard index fund (VFINX) as a proxy. Figure 1 shows a graph of the American Century Ultra fund up to September 19, 2001. and VFINX after this date (VFINX is scaled so that it dovetails with TWCUX on 9/19/01). The black arrow marks my complete movement to the income fund on 9/03/99 and the red arrow mark my subsequent repurchases of stock funds in 10% increments.

In my first article I presented I developed a series of levels (the blue lines labeled BUY in Figure 1) at which I would move 10% of my 401K from the income fund to TWCUX and series of levels (the black lines marked SELL in Figure 1) at which I would move 10% of my 401K from TWCUX to income fund. As of the time of that article (May 29, 2001), I had moved 10% from income to TWCUX five times on the dates marked in the figure and was 50% in TWCUX and 50% in income. Since that article I moved another 10% from income to TWCUX (for a total of 60%) on September 19, 2001, as I reported last April.

A new buy is shown on July 10, but now I am investing in the index fund (I was required to move all the funds from TWCUX to the index by the beginning of July and did so in June). This buy reflects the fact that VFINX was substantially lower on that date than it was in September, low enough to have reached another blue line. I am employing the same buy lines as before. For now, I will retain the same sell lines (having added another sell line).

Figure 1. Composite stock fund with buying (red arrow) and selling (black arrow) points market.

Also shown in Figure 1 is a green line which shows how my 401K has fared since Sept 3, 1999. Although I am now 48% richer than if I had stayed fully invested all through the decline, I am 19% behind where I would be if I had just stayed in the income fund. Overall my position is down 5% from 9/3/99. On the other hand, I am now 70% in stocks and should the market move upwards I will participate strongly. I plan to complete the experiment and see how it turns out, providing periodic updates as I either add positions or sell them. Without practical demonstrations of investment strategies based on secular trend ideas, the utility of the concept is unclear. The presentation in Figure 1 suggests that any reliance on stocks for my 401K is foolish. That is, I would have been better off staying fully invested in the safe income fund, as I was before 1995. In actuality this was not true, TWCUX handily outperformed the income fund over the 1995-2000 period. I do not have a record of my buys and sells before October 1997, but Figure 2 presents the total return obtained by the actual mix of stock and income funds in my 401K after 10/23/97 compared to that obtained from a 100% position in stocks or a 100% position in the income fund (assumed to be a fixed 6% return over the entire period).

Figure 2. Overall 401K return since 10/23/97 compared to stock and fixed 6% return

Having stocks contributed strongly to total return during the bull market. From 10/23/97 to the bull market peak on 3/24/00, my 401K mix advanced 43% compared to 99% for TWCUX, and only 15% for the income fund. At this time it appeared to the mix strategy in my 401K, although far better than the straight income strategy I used before 1995 was doomed to underperform the stock market. Yet as of the 7/12/02 close, the total return from 10/23/97 of my 401K mix strategy has been 35% compared to a 6% loss for the all-stock strategy and 32% for the all-income strategy. So both stocks and the safe investment options have a role to play in maximization of returns. The stock accumulation begun in October 2000 has substantially depressed my returns over what I would have gotten had I maintained a 100% income stance. Yet if I had done so, I must also consider the negative effect of holding the income fund instead of the stock fund during bull market periods. I should note that the entire period covered by Figure 2 has been a time of record stock market P/E and a 100% income stance in the late 1990's would have been in order. One can point out that if the S&P500 were to fall another 5% from present levels (~920) the total return of the mix strategy from 1997 would fall below that from the income fund. This is completely true, and as the market falls the stock allocation will increase, accelerating the erosion of total return. Should the S&P500 fall to ~700, at which point my method calls for 100% stock allocation, total return from Oct 1997 would likely fall below 10%, substantially behind the income return. In this case, the blend strategy will look decidedly inferior to the all-income strategy, just as the blend strategy looked inferior compared to the all-stock strategy in March 2000. But should that visit to 700 be followed by a strong bull market, the blend strategy can pull substantially ahead of the income strategy once again. But one can point out that by staying in the income fund until 700 is reached and then buying, a much better return will be obtained. This is true of course, but if one has stayed in income from October 1997 though the 1997-2000 bubble and then through the 50% collapse that followed the bubble, would one have the courage to go long? After all, if we use the 1929-32 bear market as a model, a 50% decline in the S&P500 could be followed by another 50% decline and the level reached would still be some 70% higher than the 1932 bottom. In other words, just as a bull market can always go higher (in the short term), a bear market can always go lower. In the real world, when faced with a powerful bear, most investors will wait until it is clear that the bear has gone before re-entering the market. And usually that means the bull market has to have recovered 50% or more of the bear market decline before it starts to look like a bull market and not just another bear market rally.