We have written somewhat extensively in the past about our Sentiment score as a measure of, well, general investor sentiment surrounding a stock, irrespective of the company’s fundamentals. In some cases we have found it to be a decent leading indicator indicative of short-term price performance, especially in cases in which the qualitative story outweighs the quantitative story.
This would be the case of, say, Netflix (NASDAQ:NFLX) shares falling as the company very publicly sorts out its streaming vs. mail delivery businesses and corresponding prices, while continuing to report very strong financial results. Its Sentiment indicator started falling in June into ‘neutral’ territory after being positive, almost without interruption, since 2009, eventually turning negative in September as shareholder and subscriber backlash over its new pricing policies reached an almost hysterical climax.
While Netflix is a widely followed story with ample media coverage, many other stocks offer significantly fewer clues about their unfolding qualitative story and yet their stock price fluctuates making one think: “somebody knows something I don’t.” Our Sentiment indicator was designed, in a way, to provide such missing guidance. As we have followed—and written about—our Sentiment indicator on individual stocks (such as our recent post on ‘10 Value Stocks with Improving Sentiment,’) we have also followed the indicator across broad groups of stocks, including sectors and industries, and have identified certain patterns that we thought would be of interest to our readers and subscribers. More specifically, we have constructed a new indicator which we have called the MarketGrader Sentiment Index, that we hope will help investors navigate today’s manic-depressive, risk-on/risk-off stock market.
The MarketGrader Sentiment Index
The new MarketGrader Sentiment Index (we’ll call it MGSI for short) simply tracks the ratio of stocks with a Positive Sentiment indicator to stocks with a negative Sentiment indicator in MarketGrader.com. Unfortunately today we only have a little less than four years of history, essentially since our Sentiment indicator was first introduced in 2008. Fortunately, however, that was four months before a 10-year market low as stocks reeled from the financial crisis. While ideally we would want to look at the index in 2006 and 2007 leading into the market decline, the data we do have today, particularly from the first quarter of 2009, offers interesting parallels with today’s market, which, as being extensively reported, is nearing bear market territory.
Today MarketGrader.com covers 6,058 North American stocks, including more than 4,600 listed on U.S. exchanges and over 1,400 in Canadian exchanges. Collectively this coverage offers a pretty broad picture of the North American stock market. Of all stocks currently under coverage, 489 now have a positive Sentiment indicator and 3,449 a negative Sentiment indicator. The result is a ratio of 0.14. Considering that in the last four years the MGSI has moved in a range from 0.13 (March 2009) to 3.41 (January 2011,) the fact that its current level is very near the all-time low should not be overlooked by those wondering where we’ll go from here. A few historical points help put these values into perspective.
As mentioned above the all-time low for the MGSI was reached right around the 2009 market bottom, specifically on March 20th, 2009. That day the S&P 500 closed at 832.86, which we now know was ten days after it reached a decade-low of 676.53 on March 9th, 2009. MGSI stayed at 0.13 for a brief week, through March 26th,before beginning a steady, almost uninterrupted climb that would peak seven months later on the back of a furious stock market rally. While the MGSI would dip into three distinct troughs after the October 2009 peak, it never breached what we now know is the important level of 0.5 again until a month and a half ago as the recent sell-off began to gain momentum (more in this later.) Please refer to the chart below for a four-year history of the MGSI plotted against the S&P 500 Index.
What happened to the S&P 500 from MGSI’s trough to peak? Between the MarketGrader Sentiment Index all-time low of 0.13 first reached on March 20th, 2009 and its peak of 3.17 first reached on October 23rd, 2009, the S&P 500 rose 40.5% from 768.54 to 1079.60. From the October 2009 peak MGSI fell for nine months to a trough of 0.69, which it first reached on July 16th 2010. In this case, from peak to trough, the S&P 500 fell to 1064.88, a decline of only 1.4%. This nine-month decline was not, however, a straight line down as has been the case this year. In between the aforementioned peak and trough the MGSI had two temporary multi-month lows and two subsequent, multi-month highs, all while the market moved sideways. Worth noting is that never during this time did the indicator climb above 3.0 or fall below 0.5. This would change, though, in the next year, bringing us to where we are today, having breached this high and low point for the first time since 2009.
Following the July 2010 low, MGSI climbed almost in a straight line, breaching the 3.0 mark on January 31st, 2011, which is now its all-time high. From bottom to top the S&P 500 rose to 1286.12, a 20.8% increase. The index stayed above 3.0 for only four days before beginning a steep, almost uninterrupted decline to its current level of 0.14, only 0.01 away from the all-time March 2009 low. In other words, today for every stock with a positive Sentiment indicator in MarketGrader.com there are seven stocks with a negative Sentiment. From top to bottom, since the January 31st top, the S&P 500 fell 14.5%.
While it might be overly simplistic, and perhaps treacherous, to call a market bottom based only on the current level of the MGSI, putting it into the context of current company valuations helps illustrate how broad and steep the recent sell-off has been. On March 7th, 2009, two days before the S&P 500 Index’s bottom, Barron’s cited, in its cover story a Citigroup report that estimated full year 2009 profits for the S&P 500 of $51, not far from the consensus estimate at the time of $52. Based on the S&P 500 Index’s close of 683.38 on March 6th, 2009, the forward multiple for the index based on 2009 estimates at the time was 13. The companies in the S&P 500 went on to earn, collectively, $62.20 in 2009 according to FactSet. Thus, with the benefit of hindsight, 13 times forward earnings was a good buying point for the S&P 500, particularly after its gut-wrenching decline of the prior year and a half. In contrast, analysts polled by FactSet Estimates today expect the S&P 500 to earn $95.40 in fiscal year 2011, putting the index’s P/E ratio at 11.5. This is not only 1.5 points lower than the March 2009 bottom multiple of 13 but it has the benefit of three quarters of reported earnings already in the books for 2011, compared to estimates in March 2009 before a single quarter of corporate earnings had been recorded. So, while the global economy might grind to a halt as buyers of U.S. government paper seem to imply today, it is pretty unlikely that the margin of error for 2011 corporate earnings would be such as to throw the current multiple off by more than a half a point or so.
But before we get ahead of ourselves it is worth looking a little further ahead into 2012. According to FactSet Estimates analysts expect, on average, the S&P 500 to earn $104.72 in 2012. If so, at today’s level, the index would be trading at 10.5 times 2012 estimates. Yes, we know, skeptics will yell: “No way, these estimates are too high!” Fair enough; if so we suggest then looking at the lowest estimate for S&P 500 earnings for both 2011 and 2012 among all firms polled by FactSet to compute their mean, which, courtesy of Jefferies, would be $82.33 in 2011 and $95.07 in 2012. Based on these estimates today’s S&P 500 would be trading at 13.4 times 2011 and 11.6 times 2012 estimates. Which brings us squarely back to the March 2009 P/E on the S&P 500 of 13 times full year estimates. Combined with the recent low in the MGSI described above, current stock market valuations spell B-U-Y.
There are a few points we would like to make in concluding. First, we understand the shortcomings of such a small data sample when referring to the MarketGrader Sentiment Index. As such expect additional research in the coming weeks on the index for the period before November 2008. Finally, while we don’t intend to track or use the MGSI as a market timing indicator, we do think there are a couple of levels worth watching for in the future, namely 3.0 on the high end (we think as an indicator of excessive bullish sentiment) and 0.5 on the low end (as an indicator of excessive pessimism.) From the brief history covered in this report both levels have only been breached twice in each case. On the high end both were followed by significant declines in the S&P 500. On the low end, the first came in March 2009, in one of the biggest buying opportunities in history. The second one, well, we’re about to find out.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.