Since 2000, I have adhered strongly to my asset allocation disciplines in moving money for clients between stocks and bonds for balanced results. It is not an exact science and as I am not a market timer - even at market extremes like March 2000 or March 2009 - I have rarely recommended being 100% in stocks or 100% in cash at significant market turns. I am a great believer in Modern Portfolio Theory which postulates that allocations to equities should almost always be in the vicinity of 20-80%.
Going below a 20% allocation to equities does not really reduce one's portfolio risk over the intermediate and long-term (it may in the short-term, of course). New risks are introduced on the fixed-income side such as reinvestment risk from money market funds and bond price risks from longer-duration instruments. Similarly, most people -- even if they know about "stocks for the long run" -- have a hard time sticking with a portfolio allocation that goes much above 75-80% allocation to equities. The risk is that without any assets offering ballast during turbulent times like 2008-09 someone too heavy in equities will bail out at or near the bottom, resulting in their portfolio being dramatically underweight stocks during the subsequent recovery.
By increasing or decreasing stock allocation to between a 25-35% allocation at the low-end and a 75-80% allocation at the high end, I have found most clients and investors are more able to stay invested in stocks and risk-assets during turbulent times, generate returns close to their required returns, and also sleep at night. The exception might be young people just starting their careers, who are earning and saving large sums relative to their portfolio, or the very wealthy who have higher risk tolerances. A few years ago I quantified this in a proprietary calculation I call the "Cash Flow Replacement Ratio" which assigns risk-weights to the proper stock/bond allocation for an individual based on years to retirement, savings level, and portfolio size. The underlying premise is that as your portfolio increases over time and as your years of working (saving) approach their end, you should be more conservative in your portfolio allocation. The old "100 Minus Your Age In Stocks" was an earlier, pre-Excel variant of this formula. You can download a sample of the CFRR here and play around with the numbers for your specific financial circumstances.
Now that we have determined that asset allocation is important, where exactly are we in the current stock market cycle to help us determine that allocation based on stock fundamentals ?
I have long believed that financial assets work off extreme levels of overvaluation in a multi-decade move directly proportional to the overvaluation and excess of the earlier move. The more the run-up into the bull market peak, the longer it takes to repair the damage. Largely, this is basic math: assuming valuation levels at the start of bull markets are about the same, those which go further suffer from much higher valuation levels (unless earnings growth was much stronger to offset the price rise).
It took gold 28 years to make a new high after the price increased 20-fold within 9 years in the 1970's.(click to enlarge)
With gold in 1980 and the Japanese Nikkei 225 in 1989, each took time off for 2 decades, burning off the excesses of the previous move (the Nikkei still has not reclaimed its old highs). For similar reasons, I believe the NASDAQ Composite - which went up 300% in less than 18 months - will need at least until 2025 before it is within striking distance of a new all-time high. The greater the parabolic move up, the longer the time to repair. Consumer staples stocks like Coca-Cola (KO) and drug stocks like Pfizer (PFE) sold at P/E levels above 30x in the late-1990's, 2-3x their long-term growth rate of earnings. Coke is within striking distance of an all-time high but Pfizer is essentially at half her 1998 peak:(click to enlarge)
Fidelity Investments has noted that secular bull markets average 21 years and produce a total return of 17.2% in nominal terms and 15.9% in real terms. The market's P/E more or less doubles, from 10x at the start to 20x at the end. The average secular bear market lasts 14.5 years and has a nominal total return of 1.0% and a real return of (2.3)%. The market's P/E falls by an average of 9 points, from 20x at the start to 11x at the end. The Fidelity bear market time frame meshes with my 16-year thesis, shortening it a few quarters but essentially the same underlying thesis on valuations and time frames.
The 1929-45 bear market took 16 years following the September 1929 peak. Importantly, the low for the Dow Jones Industrial Average (DJIA) was set 3 years into the decline, in July 1932 with the DJIA down 89% to a level of 41. Note that another great buying opportunity occurred 8+ years into the secular bear market in mid-1938, though at a higher level. The market continued to recover throughout the 1940's, setting another higher low early in 1942 when the early results from World War II did not go well. The market continued to ebb-and-flow, earnings and the economy healing themselves, valuations getting cheaper, until once the war ended the market rose for 20 years from 1946-1966. A nominal new high was hit in 1972, but essentially the market churned and kept stalling out near DJIA 1,000 for 16 years. The 1,000 level was hit (or closely approached) in 1966, 1968, 1972, 1976, and 1980; not until August 1982 would the DJIA finally achieve escape velocity and leave the confines of DJIA 1,000. An 18-year bull market then ensued.
The S&P 500 (the new market barometer) made highs in March 2000 and October 2007. The market did make a slightly lower low in March 2009 than in October 2002 but the NASDAQ was 20% above her previous lows. If one takes into account the disproportionate impact of financial stocks in the S&P 500 at the October 2007 highs and subsequent meltdown, you could make the case that March 2009 was a re-test of October 2002 and a double-bottom. Note that the stock market bottoms early or mid-cycle in these 16-year bear markets (1932, 1974, 2002 or 2009), not late. So on a timing basis we appear to be 75% of the way through the 16-year bear market.
What about valuations? Earnings in the S&P 500 are substantially higher today ($102) than in 2007 ($83) or 2000 ($56), but the S&P 500 is essentially at the same peak. This is classic 'burning off the excess' and growing into valuations that has accompanied the repair of previous bubbles, including the 1920's stock market bubble. For that reason, I believe that just as stocks bottomed in 1974 (2 years after the 1972 nominal new high and 8 years after the 1966 peak) and also in 2002 or 2009 (2 and 9 years after the 2000 peak), that we have seen the low - and probably the generational low as Doug Kass has said - for this bear market.
The time from the price bottom in 1974 to the 1982 ultimate bottom was about 8 years. So far, we have the 2002 'early bottom' (2 years after the 2000 high) and the 2009 'late bottom' (9 years after the peak) following a similar pattern. Note also that the market made the 1932 DJIA low 2+ years into the decline and another important bottom was made in 1938 (9 years into the bear market). Thus, important lows are being made 2+ years and 8-9 years into all of these 16-year bear markets.
So if the previous patterns hold, it means that the market should be ready to break out in the vicinity of 2016. The market will probably have at least one and perhaps two serious corrections before then. The good news is it is highly unlikely we will approach either the October 2002 or March 2009 lows, barring stupidity on the part of elected officials (always a possibility). If the pattern of higher-lows seen in 1929-45 and 1966-82 holds, then something in the vicinity of S&P 1,100 (the Euro Crisis 2011 lows) would likely be the worst-case level for the S&P 500, unless we have a severe Lehman-type policy accident in the U.S., Europe, or China. That would be a 30% correction from current levels. S&P 1,250 (a popular support/resistance level over the years) would represent a 20% correction and S&P 1,350 (last November's lows) would be a 13% drop from today's levels.
Besides the valuation and timing factors finally helping equities, the demographic headwinds noted by many Seeking Alpha commentators will no longer be a negative for the stock market and will in fact be neutral-to-positive out to 2030. The middle aged-to-young ratio (MY ratio) will be increasing:(click to enlarge)
As with any other indicator or variable, demographics are not 100% predictive or fail-proof, but lining up with other timing and valuation measures improves the odds for a lasting bull market, the kind we had from 1982-2000. All of the headwinds turn into tailwinds. If P/E ratios are going lower over the next few years that can be accomplished by a stock market decline that is excessive relative to an earnings drop, perhaps a U.S. debt or Euro problem. Without an earnings collapse, stock prices need not collapse. As the time cycle stretches out a few more years, if stocks trade in an S&P 1,250 - 1,600 range but earnings strengthen, we 'burn off' more of the valuation excess that began in 2000. The market can do anything it wants - perhaps March 2009 was the start of a new secular bull market - or it can bottom anytime between now and whenever. But a re-test of S&P 1,100 or the other targets cited previously with earning that fall much less will make the market even more cheap on trailing, forward, and Shiller Cyclically Adjusted P/E levels:(click to enlarge)
In 2011, the market fell 20% during the Euro & Debt Crisis and earnings were hardly impacted. Another factor in the ability of the U.S. market's ability to stage a multi-decade bull market run is the increased share of S&P 500 profits that are coming from overseas. American companies are increasingly like Japanese auto and electronic manufacturers in the 1970's and 1980's, exporting to the world. In the 1950's only 5% of the S&P 500 earnings came from overseas, a figure that rose to 20% in the 1980's, 30% in the 1990's, and is over 40% today. By 2020 over half of the profits of the S&P 500 are expected to come from outside the United States. As globalization has increased, U.S. corporations have been able to use their operating leverage to increase profits globally.
Two well-regarded forecasters, the Leuthold Group and Jeremy Grantham of GMO, both see low-single digit returns from stocks over the next 7 years from current levels. With those 7-year forecasted returns, it is unlikely we are at a launching point like the absolute lows of 1932 or 1974, or the absolute end of the bear markets in 1946 and 1982. We may have just passed in 2009 the equivalent of 1938 and 1978, steep corrections that produced buyable bottoms. Had Lehman Brothers (a policy accident) not been allowed to fail, the market probably would never have gone down as much as it did in 2009 and 2002 would still be this bear market's low. Remember, the market recovered nicely in early-2008 when Bear Stearns was mopped up by JP Morgan and the Federal Reserve, unlike the situation with Lehman Brothers.
Cheaper valuations on stocks -- and higher rates on bonds -- are both required because right now the traditional 60/40 stock-bond portfolio is projecting a very low real rate of return. This is a result of rock-bottom interest rates and only "OK" valuations on stocks:(click to enlarge)
To sum up: stocks are likely a bit ahead of themselves right here, considering policy and political risk both domestically and internationally. Stocks are not at excessive valuations, but they are not cheap, either. A continued pattern of growth in the United States and globally disproportionately benefits large corporations like those in the S&P 500, where unique American brands (Coke, Philip Morris, 3M) and technological advantages (Cisco, Apple, IBM, Microsoft) are allowing the U.S. to become increasingly export-oriented to faster-growing economies. This will help to preserve and possibly maintain the long-term 7% earnings growth rate for the S&P 500 since 1960.
If earnings continue on that trajectory for another 3-4 years and the market stays at current levels or declines sometime in the next few years (ideally, 2016) then believe it -- or not -- you are looking at a situation where timing cycles, valuations, demographics, and global GDP growth are all lined up to propel the S&P 500 to 4,000 by 2030.
16 years is a long time to wait, but that would be something worth waiting for !