Market Timing

By Ray Schumacher

Note: Text Only. Click HERE for a full version with all figures, charts and tables. PDF Reader required.

Style Index Investing (SII) makes no mention of Market Timing. Indeed SII is based on holding on to one style index fund or another through good times and bad. SII acknowledged that this is more easily said than done, given the psychological pressure to follow the crowd. But as SII pointed out, during the Extended Market Downturns of 1981 and half of 1982 and 2000-2002, it cut losses to 46% and 17% respectively to what they would have been had you held a S&P 500 Index Fund.

 How nice it would have been to have avoided these losses altogether.

Successful Market Timing has always held great allure but everything I read on the subject said it couldn’t be done. The main objection was that you had to be right on both when to get out and when to get back in, and because so much of market movement is governed by human psychology, the chances of being right at both ends were slim to negligible.

Being a skeptic, I thought this might be another one of these Wall Street myths that don’t hold up under close scrutiny. I reasoned that because the only widely discussed measure of over or under pricing of stocks was the Price Earnings Ratio (P/E), it covered only half the story. What about the attractiveness of the alternatives to stocks?

 What kind of returns do Bonds or Money Markets offer as compared to stocks?

One needs to relate both of these variables together in some way to truly judge whether stocks are over or under priced, and if this is done, maybe Market Timing could be made successful. A clue as to how to do this came to my attention.

Somewhere in the past, a formula was devised to relate stock valuations with bond yields. It is called the Federal Reserve formula. Whether it was devised by the Fed or used by them, I do not know. Regardless, it made some sense to me.

Basically it says that if the ratio of the Expected Earnings Yield on stocks (S&P 500) to the 10 year Bond yield is above 1.00, then stocks are the better buy. Under 1.00, Bonds are better. The earnings yield on stocks is merely the reciprocal of the P/E.

I saw two problems with the formula. The earnings yield is based on the expected earnings of the S&P 500 companies on the theory that market behavior is based more on expectations than on what has actually happened recently. My problem with this was that expected earnings were always changing, seemed to err on the high side, and were fraught with so-called adjustments that I suspected (and in many cases, later confirmed) were excuses for past management mistakes. The remedy was to deal only with actual reported unadjusted earnings.

The other problem concerned the 10 year Bond Yield. Bonds are a fairly safe investment if held to maturity, but with a 10 year life, and knowing that bond prices go down when interest rates go up, and that periods of owning them would likely be much less than 10 years, it seemed obvious that the 10 year bond yield is not the right number for the denominator of the ratio.

This is confirmed by the fact that cash or very short term debt obligations are the alternative used by most stock mutual funds. Thus, in place of the 10 year bond yield, the current money market rate seems more appropriate. Actually, I use the 90 day Treasury Bill Discount Rate which is set by the Fed and is the basis for all short term borrowings of both the U.S. Government and businesses, including banks.

 (See full complete report for data)

A history of this ratio starting in 1979 is shown in Graph 1. As you can see, there are wide swings above and below 1.00 with very low ratios as we approached the bubble burst in 2000 and then extremely high ratios thereafter as the Fed tried to repair the damage by lowering the Federal Funds Rate (essentially the same as the 90 day T-Bill discount rate) to unprecedented lows. During 2003-2004, the results are off the graph scale. Since 2004 as the S&P 500 has recovered and the Fed has raised rates, the ratio has decreased but is still high compared to past norms. (1.2 as of April, 2006).

So as the S&P 500 P.E. increases, the S&P 500 earnings yield decreases. This results in the stock/cash ratio decreasing (assuming that the 90 day T-Bill rate remains constant or changes at a slower rate).

Graph 2 shows the S&P 500 with dividends included over the same period. The path is somewhat smoother, but close inspection will show the decline during 1981 through mid-1982, the “crash” of 1987, the recession of 1990, the Indonesian Currency Crisis of 1998 and of course the major decline from 2000 through most of 2002.

Now let’s see if the ratios of Graph 1 have any relation to the declines and recoveries of Graph 2. First we need to recognize that human psychology plays a major role in changing the direction of the market. The momentum of good times carries beyond the point of rational change, so that the ratio can decline well below 1.00 before the market turns down. Similarly, the momentum of bad times and the fear it generates requires the ratio to move well above 1.00 before the market turns up again.

Assuming the effects of euphoria and despair are equal, we may set boundaries equidistant above and below 1.00 to mark the ratio levels below and above which we will switch to cash or back to stock. The data indicates these to be 15 percent below and above 1.00, or .85 and 1.15.

Using these limits, the switch dates are as follows:

Table 1 summarizes the results, showing that over 26 years $1.00 grew to $28.28 by holding the S&P 500 and to $47.15 by switching between it and the Money Market. The advantage of Market Timing (switching) compounded annually was 2.2% per year. It is important to note the far right column of Table 1 where the advantage of switching out of Stocks is shown for each of the 5 periods that the .85/1.15 rule said to be out of Stocks.

The last 2 periods (6/1/90 – 1/1/92 and 8/1/97 – 9/1/01) actually yielded negative results because in the case of the former the ratio was a year late in getting above 1.15 and in the case of the later, the ratio had us out of Stocks way too soon because euphoria perpetuated the rise for almost 3 years beyond 8/1/97 when the .85 limit said it should end. The bubble finally burst in Early 2000.

Human psychology is hard to predict. The modest success of Market Timing in this case is attributed to the 3 good signals outweighing the 2 bad ones. How will it work in the future? It’s hard to say. Bubbles like the recent one are fairly rare (1929 and 1972 come to mind) but the late signal to get back into the Stocks after the 6/1/90 signal to get out could happen again at anytime.

Fortunately the Style Index Investor doesn’t have to worry about it. Table 2 (Next Page) is just like Table 1 in that switch signals are the same, but Style Index performance is substituted for the S&P 500. In this case, performance between the indices and the Money Market yielded a negative advantage of 1.1% per year with $1.00 growing to only $127.39 for switching over 26 years versus $158.33 for not switching.

Again note 3 of the 5 switches to cash were profitable while the other 2 were not, and by a very wide margin in the case of the 8/1/97 switch to cash. (11 months) until the bubble burst, and then to Small Cap Value Stocks where investor preference remains to this day (April 2005).

One might say: Well all of that is fine, but I’m not sure I want to stay fully invested during future extended Market Declines, regardless of the fact that Style Index Investing will cut my paper losses. If I’m retired, I’d have to sell some of my Style Index Portfolio to cover living expenses and as a result convert paper losses into real losses. This would be painful. But on the other hand, I hate to lose 1.1% per year of the overall Style Index Advantage. Isn’t there some way to avoid turning these paper losses into real losses by settling for some reduction in the advantage but not as much as 1.1% per year?

 The answer is Yes.

Let’s say you are retired and withdrawing from your portfolio at a rate of 5% per year to cover living expenses. When the ratio (Schumacher version of the Federal Reserve Formula) drops below .85, you transfer 5% of your portfolio to the Money Market to cover one year of living expenses. Below .75 you transfer another 5%, and below .65 another 5%. When the ratio rises again above 1.15, anything left in your Money Market account is transferred back into the favored Style Index Fund.

You have, in effect, bought an insurance policy against turning paper losses into real losses, and probably gained some restful sleep. This comes at a cost of .2% per year as shown in Table 3.

As you will note in the far right hand column, this cost accrued mainly during 1998 and 1999 when despite ratios under .75 the Market continued its rise. At that time you might have questioned this whole idea of insurance but when the bubble finally burst in early 2000, it would have seemed like money well spent. Again, those with nerves of steel can avoid paying the .2% per year insurance premium, but for the average investor, it might not be a bad idea.

This publication is not to be summarized, reproduced, or rebroadcast in any fashion without our written permission. Nothing herein should be construed as an offer, solicitation or recommendation to buy or sell any security. Style Index Investing does not accept compensation from any mutual fund companies. Past performance does not guarantee or imply future results. We are not registered investment advisers. Consult your own financial advisor before investing any money!

Note: Text Only. Click HEREfor a full version with all figures, charts and tables. PDF Reader required.