I'm not sure if 2013 will go down in history for any major market moves in the S&P 500, although the run to date has been rather robust. However, based on the more conservative PE ratio using a 10-year cycle, recent valuations should give investors at minimum some pause.
The Cyclically Adjusted 10-year Average Price-to-Earnings [CAPE] continues to rise, and is now at its highest mark of 23.9X since the S&P 500 bottom of 13.3X set back in March of 2009.
Who cares right? CAPE is not a market timing tool and why would it matter to those who invest in stocks - not the market? While I am not here to be the arbiter of proper valuation tools or their use, I do consider some history related to the current market environment.
There are only 4 times going back to the 1920s that the CAPE was at or above 24X. During those times the 12-month trailing earnings were running significantly above earnings 10-year average. During three of those periods, the 10-year average was running above the 85+ year trend with the 1999 period having just reached the long-term trend.
Figure 1: Shiller CAPE 10
Figure 2: S&P 500 Earnings 1-Year vs. 10-Year Average
Figure 3: 1-Year Earnings percentage Above or Below the 10-Year Average
The following looks at the 4 periods in more detail.
The last time CAPE peaked at the current level and turned down was in 1966 at 24.1X. At that time, the 12-month trailing earnings were 35% above the long-term trend. The average is 8% above trend given that the trend is rising. At that time, the 10-year Treasury was 4.6%.
The CAPE ultimately troughed in 1982 at 6.6X, and interest rates were at 14.0% (having peaked some 9 months earlier at 15.3%). The trend in 12-month reported earnings at that time were 15.2% below the 10-year average and falling. Ultimately earnings would trough in 1983 at 24.6% below trend.
Going back to 1929, CAPE peaked at 32.6X. At that time, the 12-month trailing earnings were 61.5% above the 10-year average. 10-year interest rates were at 3.4%. CAPE subsequently troughed in 1932 at 5.6X earnings. Trailing 12-month earnings were 40.5% below the 10-year average at the CAPE bottom. Earnings troughed in early 1933 at 50.4% below the 10-year average. There was essentially no change in the nominal interest rate over that time period, although the real interest rate net of inflation (using the CPI data provided by Professor Shiller) did jump to 12.8% in mid-1932 from 2.2% at the CAPE peak.
In December of 1999, CAPE multiple reached 44.2X. Earnings at the time were running 49% above long-term trend. During this period, the strong growth in earnings from the December 1991 trough brought the 10-year average back to its long-term trend. Interest rates in late 1999 were at 6% with modest inflation.
The CAPE troughed in 2003 at 21.2X. Trailing 12-month earnings at that point were 25.5% below the 10-year average. This period was slightly different in that earnings troughed in late 2001 and early 2002 - which was before the CAPE low. 12-month earnings dropped to 35% below the 10-year average in March of 2002.
This period is somewhat unique that the sheer magnitude in the CAPE peak during the dotcom euphoria. With an earnings recovery, valuations were sustained at higher levels than the other 3 periods discussed but ultimately gave way in 2007/2007 period discussed next.
The 2007 time period is unique, given that it saw a plateau of CAPE valuations above 24X from June of 2003 to January of 2008. The actual high water mark was set in May of 2007 at 27.6X. At that time trailing earnings were running 54.3% above the 10-year trend.
The CAPE troughed at 13.3X in March 2009. Given the significant write-offs during that period, the trailing 12-month earnings dropped to a massive 87.9% below the 10-year average.
The yield on the 10-year Treasury Interest declined roughly 200 basis points (4.8% to 2.8%) over that time period, although real yields jumped from 2.8% to 5.6% (based on y-o-y change in the Consumer Price Index).
In context of a secular bear market starting in 1999/2000, the 2003-2007 valuation period was an interim before another leg downwards based on the earnings cycle.
The S&P 500 valuation based on the CAPE has been creeping towards 24X. While this alone is not a catalyst for any change in market direction - the combination with above-trend earnings could. In other words, it is the combination of a high CAPE valuation and above-trend earnings that creates an unfavorable market environment.
However, unlike the previous episodes (excluding the plateau of 2003-2007), the recent market has been one of a rising market valuation while real earnings growth has stalled and is currently declining (through March). As we saw in the 2003-2007 period, it could be that it will require the completion of the current business cycle which started in June of 2009 to see valuations reach lows seen in previous time periods discussed above.
Still, it is tough to imagine that the divergence in earnings and valuation will be sustained. Either earnings growth will rebound or market valuations will compress. Current market expectations are for a strong rebound in earnings for the second half of 2013. This second half rebound is expected to reflect earnings growth of ~19% for full year 2013 and net of expected inflation of ~2%, real earnings growth of ~17% (based on S&P data dated 7/8). Robust earnings growth however is a lower probability outcome in my opinion.
There is likely a higher probability of earnings revisions downwards for 2013 based on the classic reversion to the mean of current earnings compared to the 10-year average (and the large ? label in figure 3). Almost a year ago, we wrote in a note that there was a greater chance that the trend in earnings growth would revert to the mean long-term growth rate by declining, which was contrary to positive earnings expectations. It turns out that 2012 real earnings growth declined 2.2% versus expectations at the time of 3+% real earnings growth. We see a similar situation in 2013, with real earnings growth not living up to expectations. We will know more after the next two weeks when a majority of companies in the S&P 500 report 2nd quarter earnings.
While history never repeats itself, the current market environment of a CAPE nearing 24X and earnings above long-term trend averages should reflect caution based on similarities to past periods. Furthermore, while the trend in earnings is expected to rebound in 2013, there is a likelihood of lower growth expectations based on the reversion to the mean of trailing 12-month earnings to the 10-year average and to a lesser degree the 10-year average relative to the 85+ year long-term trend.
I remain invested in this market and in the S&P (and large cap stocks in general) but despite the underperformance in other asset classes, I wouldn't recommend overweighting the S&P 500 or domestic large cap stocks in an attempt to improve performance.