TIME TO GET SENTIMENTAL

In mid- September 2014, the media and several observers discoursed on the apparent disappearance of the bear specie (NO MORE BEARS! THE BARE FACTS). Don’t worry, they were only hibernating. They are waking up, scaring the heck out of the bulls which are soon to be pronounced near-extinct by President Obama as this chart from Ed Yardeni shows:

The last time their were so few bulls was at the end of 2008, even though there were substantially more bears then. What we are having a lot now are baby bears with a record number of investors in the “correction camp”, more than twice as many as in 2008.

The bull/bear ratio has the best record as a contrary indicator. Even though the number of true bears is not as high as I would like, the large cohort of baby bears pleases the contrarian in me and makes me take the current bull/bear ratio seriously.

Also of interest is the current low equity exposure of the AAII members, nearly 40% of whom are bearish (+11% in one week!) (chart from The Blog of HORAN Capital Advisors)

The S&P 500 having corrected 12% from its May high to its August low close, investors are yearning for direction and the talking heads and other pundits are obliging with scores of economic scenarios. Indeed, this is a very complex world with many possible outcomes as the Fed is realizing.

EMPEROR: My dear fellow, there are in fact only so many notes the ear can hear in the course of an evening. I think I’m right in saying that, aren’t I, Court Composer?
SALIERI: Yes! yes! er, on the whole, yes, Majesty.
MOZART: But this is absurd!
EMPEROR: My dear, young man, don’t take it too hard. Your work is ingenious. It’s quality work. And there are simply too many notes, that’s all. Cut a few and it will be perfect.
MOZART: Which few did you have in mind, Majesty?
EMPEROR: Well. There it is.

Unlike most strategists who base their market views on what they expect from the economy, then hope that their economic scenario unfolding (?), this will translate into the desired course for equity markets, I prefer to first analyse equity markets, check if valuations are reasonable or not, then see if there is, or are, economic scenarios which can support the appropriate stance on equities.

Economic forecasting is a highly inexact science. It is thus sensible to begin with the more “exact” analysis of equity valuations. When valuations appear excessive, it is advisable to exert patience and discipline, perhaps work on your golf game or go fishing for a while, rather than attempt to scenarize something which might make the impossible possible, like most everybody else is doing in these circumstances.

At the opposite, when valuations seem appealing and calculated risk is low, the golf game can wait and angling turns to bottom fishing while one waits for a credible scenario to materialize and help unleash values.

Looking at equity markets, one can analyse scores of data and relationships. But, in truth, it all comes down to earnings and earnings multiples.

YEARNINGS

There are 3 major aggregators of corporate earnings and this is one of the few times when their respective methodology produce significantly different results. Thomson Reuters and Factset have similar “more liberal” methodologies while S&P is more GAAP-observant. S&P digs through every company report and weeds out “non-operating items” in order to produce consistent and coherent results across industries and time. S&P is generally the best source for truly operating results.

This time, however, S&P data for the S&P 500 Index might be too conservative.

S&P treats commodity-producing asset write-offs as operating expenses, which they are in reality since these companies are in fact investing in producing assets whose values vary greatly according to the assumptions on the eventual selling prices for the commodity. When oil prices fall significantly like they have in the past year, there is a meaningful asset value impairment which must be reflected in GAAP income statements.

Because these write-offs tend to be one-offs, most investors look beyond the charges, generally disregard the one-time events and concentrate on “recurring” earnings, even though the “one-time-event” might recur if the price assumptions prove to be still too high.

S&P-calculated EPS for the S&P 500 Index meaningfully diverged from other aggregators’ in Q4’14 (-12.4%), continuing in Q1’15 (-9.8%) and in Q2’15 (-12.9%). As a result, trailing 12-m EPS are now trending very differently depending on which tally one uses.

Such discrepancies are becoming important as equity markets correct since trailing 12-m EPS per S&P ($107.31) are 10% lower than Thomson Reuters’ and Factset’s ($119.27). At 1950, using S&P EPS, the absolute P/E for the S&P 500 Index is still high at 18.2 while the Rule of 20 P/E is 20.0, right on fair value. Using the higher earnings as per TR and Factset, the P/E declines to 16.3 and the Rule of 20 P/E drops to 18.1, both much more attractive levels from a risk/reward viewpoint. In fact, 18 on the Rule of 20 is  where the S&P 500 bottomed out in mid-October 2014.

Importantly, trailing EPS are still rising per TR and Factset, providing a more solid underpinning to equities, even more so with inflation stable at its current low level. Factset calculates that if the Energy sector is excluded, the blended earnings growth rate for the S&P 500 was +5.9% in Q2 compared with the -0.7% decline in total EPS. Q3 EPS are currently expected to decline 5.1% but excluding Energy they would advance 2.3% if current estimates are met.

Corporate guidance on Q3 results have not deteriorated overall…

…while services companies have generally raised guidance.

Since 2009, there have been 3 periods when S&P-calculated EPS were significantly below its competitors’. In Q4’12 and Q4’14, S&P treated large pension liability adjustments by a few large corporations as operating expenses. In both quarters, the effect on trailing 12-m EPS was in the 5-6% range and sophisticated investors knew that only a few companies were responsible for the gap and that these would be one-off events. In the last 2 quarters, write-offs in the Energy complex made the series diverge again but, this time, Energy companies (and a few Materials companies) caused the discrepancies to the point where TTM EPS are now 10% higher per TR and Factset.

Throughout 2008 and the first half of 2009, S&P earnings were also lower than others’ as losses from the bursting of the housing bubble erupted. S&P trailing EPS troughed at $39.61 in Q3’09 while TR’s troughed at $50.84, 28% higher. When the S&P 500 Index bottomed at 666 in early March 2009, the trailing P/E was a still high 16.8 per S&P but a rather low 13.0 per TR. My own work at that time (S&P 500 P/E Ratio at Troughs: A Detailed Analysis of the Past 80 Years) suggested that trough EPS were in fact $57.74 in Q4’08 (trailing P/E of 11.5x and Rule of 20 P/E of 10) so that

• Valuation using the Rule of 20 method gave “trough” valuation of 791-923 for the S&P Index using trailing earnings, 20-40% above the Index value.
• Using 2009 “recurring operating earnings” estimates, “trough” valuation would be 720-840, 10-25% above the Index value.

…and that, accordingly, the risk/reward balance was strongly favourable to investors. Actually, these valuations were generational lows, as we all well know, now.

This is not a bear market environment: no recession in sight, no inflation to be addressed by the Fed, no serious excesses needing radical remedies and, importantly, no hugely excessive valuation levels.

This is a valuation correction like that of the spring 2010 (-13%), the summer of 2011 (-15%) and October 2014 (-10%). The former brought the Rule of 20 P/E from 19.0 to 15.5, the next one from 18.7 to 16.3 and the latter from 20.0 to 17.9.

At 1950, the S&P 500 Index sells at 18.0 on the Rule of 20 P/E which restores the risk/reward ratio into positive territory for investors, as long as earnings stay up and/or inflation stays low. In effect, the Rule of 20 P/E seldom gets below 17. This provides a good valuation floor to equities (-5.5%). On the upper end, fair value of 20, where valuations tend to over time, is 13% higher with current earnings, allowing for more upside if economic conditions allow earnings to rise further.

This is the Rule of 20 chart using the higher TR trailing EPS. Fair value is 2170 providing 11% upside against 7% downside (1812) upon a decline to 17.0 on the Rule of 20.

Here’s the same chart using S&P’s lower EPS. Fair value is 1953 providing 0% upside against 17% downside (1626) upon a decline to 17.0 on the Rule of 20.

This is spooky October and most everybody is confused as to the U.S. economy, the Fed, China, etc. Volatility is high and the 200-day m.a. on the S&P 500 has turned down. Yet, I have confidence that the U.S. consumer (IS THE FED IN LEFT FIELD, AGAIN?) will spend during the coming holidays, that inflation will remain subdued and that profits will hold ex-commodities. Going back to 3 stars.