Progress Update on the Current Secular Bear Market
Recently, I wrote an article describing the progress of the current secular bear market compared to previous ones. Events have moved so quickly since then that another update is in order. In that article I wrote this:
A caveat to this analysis should be noted. Whereas the drop in P/R of 40% is consistent with the three previous non-Depression secular bear markets (average 38% drop) the absolute level of P/R at 0.9 is higher than the value at the bottom of the first bear market for these same secular bear markets (average 0.76). That is, if the relative decline in P/R is what repeats, we could be at the bottom of the first bear market now, whereas if the absolute level is what repeats, then a lower level of P/R (consistent with an S&P500 in the low 800's) would be more consistent with history.
As of last Friday (July 19) P/R has fallen to 0.77, making the progress of this bear market nearly equivalent in P/R terms to the average bottom of the three non-Depression secular bear markets. Figure 1 shows an update of the P/R graph. Friday's decline also triggered another asset allocation change in my 401K, bringing it up to 80% stock.
Figure 1. P/R for previous secular bear market periods
Using Standard and Poor's earnings estimates, the trailing 12 month earnings on the S&P500 index is 28.37, which at Friday's close of 848, yields a P/E of 29.9. Prior to the late 1990's bubble the highest P/E values on the index and its precursors were in the 26-27 range on three previous occasions: 1894, 1933 and 1992. In each case the high multiple resulted from a precipitous drop in earnings. Another 10% drop in the market would put the current market multiple right in line with these previous incidents of high P/E due to falling earnings. I note that over the five years after 1894, 1933 and 1992 stocks gave annual returns of 11%, 6.7% and 16%, respectively. If we consider only the first two examples (the third occurred within a secular bull market and might not be relevant), history has shown average returns of close to 9% over the five years following periods like this. A statistical analysis in my previous update article using P/R suggested a median expected return of some 8% from periods like this.
Having discussed what the first bear market of a secular bear era looks like on average, a question naturally occurs. What if this bear market is not average? That is, how low can we go? When I discussed my 401K asset allocation strategy in an article in May 2001, I described how I developed a reasonable estimate for the worst-case bottom for the bear market:
I needed to ascertain where the bottom of my scale buying would end up. That is, how low would the market have to go for me to be 100% invested in stocks? In chapter 4 of Stock Cycles I presented a model I developed that described the behavior of the market over shorter periods of time. This model suggested that as long as earnings growth continued unabated (that is, we would have no recession) the stock market would continue climbing. The very optimistic forecasts of Dow 36,000 and higher made by various bullish analysts would actually come true if the ten-year old expansion keep right on going for another 7-10 years. If it came to an end there would instead be a rather severe bear market, which would very likely mark the beginning of a secular bear market predicted by P/R.
In Stock Cycles I used the model to analyze a hypothetical recession that started with an earnings peak at the end of 1999. The model I developed is not predictive, one of the inputs into the model is the index value, so to use the model one must know what the market will do. Of course one doesn't know the future path the market will take. What one can do is put in various hypothetical market scenarios into the model and see if they are consistent with past behavior. So I inputted a hypothetical recession by using the last recession as a template. By doing this I found that at a recession market bottom the S&P500 would have to be in a range of 650 to 1360 for it to be consistent with past behavior. So the worst case for a recession beginning at an earnings peak at the end of 1999 would be about 650 on the S&P500-assuming a recession of 1990 severity. Since earnings did not peak at the end of 1999, they would necessarily peak higher and so I picked 700 as a bottom.
We have not yet reached 700 on the S&P500, and my allocation has not yet reached 100%, having gone to 80% last Friday. The worst-case scenario for which I had planned has yet to materialize, but it now looks distinctly possible. The 700 estimate was developed based on my analysis in Stock Cycles in which I assumed a repeat of the 1990 recession as the driver for this bear market. The fundamental data on earnings and interest rates are in now. We now know that earnings-wise this bear market was distinctly worse than that in 1990.
Figure 2 shows a plot of the market model developed in Stock Cycles showing the period after Dec 1999. (This figure is analogous to Figure 4.10 in Stock Cycles.) In Stock Cycles (writing in early 2000), I employed a hypothetical bear market using the 1990 recession and bear market as a model for an anticipated 2000 recession and bear market. I then compared my assumed 2000 bear to the 1990 bear to in terms of model predictions. In general, during bear markets characterized by earnings declines, the actual value of the index is typically considerably higher than what the model predicts. I call the positive discrepancy between what the model predicts and what the index actual does a rational premium. Rational premiums occur because the market anticipates earnings growth in the recovery and so stocks are not fully discounted for lower earnings associated with the bear market/recession.
For example, at the end of 1991, the actual value of the index was much higher than what the model predicted (see Figure 2). This premium can be justified by plugging in a reasonable estimate for future earnings into the model for the conditions prevalent in late 1991. By extrapolation of the long-term earnings growth trend to the mid 1990s, a projected value of 30 for the average earnings in the 1990s was obtained (the actual value was 29.3). When S&P500 earnings of 30 was plugged into the model for December 1991, a value of 450 was obtained, 16% higher than the actual index level in 12/91. Thus, since the future was projected to have higher stock prices, it was reasonable for the market to be valued as high as it was in late 1991. The same sort of argument can be used to explain the high P/E seen in 1894, and 1933 as well as that in 1992.
Figure 2. Market model compared to index (Figure 4.10 from Stock
I then went on to do the same analysis for the projected bear market that was to start in 2000. I inputted an estimated average earnings of 55 for the 2000's decade and obtained two estimates for future market levels: 650 and 1360. A market completely stripped of the late 1990's mania would likely be worth 650 going forward, while one that maintained all of the manic bullishness would likely be worth 1360. Since the market was already higher than 1360 in 2000 it would impossible for full bullishness to be maintained in the event of a bear market. So I expected the bear market to fall well below 1360, and if psychology got sufficiently negative, fall as low as 650. Last year, with an extra year of earnings gained, I adjusted the bottom to 700 and devised by 401K asset allocation strategy to reach 100% stock at around 700 on the S&P500.
But now we have the actual record of earnings, interest rates and stock prices for this bear market, and actual future estimates can be made using real data and not a proxy. When I put the same estimate of 55 for average earnings in the 2000's into the model under the present conditions, I obtained 804 and 1792. That is, were the full levels of manic bullishness of the late 1990's still present today, the market would anticipate an index value close to 1800 just a few years from now. On the other hand, if the late 1990's mania was completely purged from the market, then the market would anticipate a level around 800 in the next few years, which is lower than the level today.
If we discount three years of a 5% "safe" rate from this 800 level, we get a value of about 690 for the worst-case value for the bottom of this decline. Thus, my original estimate of 700 was pretty good. This analysis assumes, of course that valuation remains rational. The difference between the projected future values of 800 versus 1800 is based on whether the market takes on a discounted earnings view of the future (psychologically bullish, or even manic near the ends of expansions) or whether it takes a "current earnings" view of the future (cautious, but not overtly bearish). It is possible that the market could take an overtly bearish view of the future, in which case the S&P500 could go down to the model value of about 400 or even lower. This has happened only once before at an early stage in a secular bear market era (although it is commonly the case late in the era). This exception was in 1932, and the extreme bearishness of the time reflected the disaster of the Depression. As I describe in The Kondratiev Cycle and here, I do not expect a Depression and so I rule out this possibility.
The main thing to remember at this juncture is that, as low as the market has gone, it is nothing out of the ordinary for a secular bear market. The market is still operating within the normal parameters I forecast more than two years ago. That is, the market has not gone anywhere near low enough to be "predicting" a Depression. Were a Depression on the way I would expect the S&P500 to be already below 500 and heading down rapidly.
That said, we did not have to go as low as we have, obviously I would not have started buying in my 401K as early as I did had it been a sure thing that we would see 850 (or lower) during this bear market. But, I was aware that we could go this low (and lower still) and it not be grossly abnormal. Time will tell how these analyses turn out, the purpose for this articles is to make the projections ahead of time to see how accurately (or not) they reflect what is to come.