The bulk of third-quarter earnings are now in. Here are three observations that have been widely discussed: earnings fell about 15 percent from this time last year, which was better than expected at the beginning of the reporting season. More than 80 percent of companies announced earnings above analyst's expectations, but only about 55 percent of companies beat on revenue expectations. The bulk of the earnings declines were delivered by economically sensitive groups like materials and energy stocks, while the financials posted the largest gains in operating earnings. Here are three more observations which have received less attention: some amount of investor selectivity or "discretion" is returning to the stock market. Higher quality companies may be staging a comeback. The earnings surprise beat rate may be useful, but not in a way that you might expect.

It's important to recognize that analyst estimates for say, year-ahead earnings, are almost always far too optimistic. As the actual reporting period draws closer, analyst estimates for various companies are gradually cut (as was the case prior to the third quarter reports). So when we say that reported earnings were generally above analyst expectations last quarter, that fact comes with the caveat that those expectations were the whittled-down expectations that existed just before the actual earnings were announced. Of course, the other caveat is that an earnings report qualifies as an upward surprise even if it surpasses analyst expectations by just a penny, which is common. For that reason, the average earnings surprise or disappointment prompts the market to reward or punish the stock price only by something on the order of 2% or less over a given quarter.

Before talking about the comeback of investor discretion, it's important to take a quick look at one example of investors' lack of discretion earlier this year. As the market moved higher this year, some observers claimed that companies were reporting results that were better than expected, and these results were fueling the stock-market recovery. When you look at the data, this argument isn't persuasive.

The bar chart below shows the average net performance of stocks in different groups. The three bars on the top show the net performance of stocks that reported negative surprises in both earnings and sales, and then in each earnings and in sales. The three bars on the bottom show the net performance of stocks using the same metrics, but include only those that surpassed estimates. The return periods run between each company's reporting dates. So, for example, Alcoa's returns run from July 8 th through October 7 th of this year, following the company's second-quarter report. The returns are market adjusted (versus the return of the S&P 500 equal-weighted index), and then averaged. Generating returns in this way allows us to focus on what investors are responding to on a company-by-company basis in the periods following the release of new information.

Although top-line sales growth is getting more attention recently, typically investors are more sensitive to earnings surprises. And as the graph above shows, investors did respond to earnings, just not in the way that is typical. Companies that announce earnings surprises tend to outperform companies that miss estimates, on average. But during the second quarter reporting season, that trend was mostly reversed. Companies that missed estimates performed best. And although it's not shown, the performance during the first quarter reporting season was similar, especially in regard to earnings surprises. Companies that announced positive surprises trailed the market by about 2 percentage points, on average. Those companies that missed estimates outperformed the market by about 2 percentage points.

The next chart updates the performance of stocks during the third-quarter earnings reporting season. Here, company performance was measured from the company's most recent earnings announcement date through Friday. You can see that there are some signs that discretion is coming back into the market. Companies that reported positive surprises have outperformed the market - and the better the news, the better the performance has been. Companies that reported positive earnings surprises are doing better than companies that reported only positive sales surprises. And companies that have reported better-than-expected results in both measures have outperformed companies that surprised on just one metric.

The rally up through late summer was less about improving corporate performance than it was about reversion to the mean and beta chasing. Investors mostly indiscriminately bought highly cyclical and highly levered stocks, partly because some of these stocks were priced for potential bankruptcy and partly based on the view that the economy will recover strongly next year. More recently, investors are showing a bit more discretion, by seeking out companies that are producing sales and earnings surprises in a very challenging economic environment.

A Turn for High Quality Stocks?

The beta-chasing pursued by investors this year created stark differences in the performance among individual stocks. One of the clearest places to see this was the difference in performance between lower-ranked companies based on measures of quality (mostly the dependability of operating results) and higher-ranked companies. High-quality companies have returned about half what the equal-weighted S&P 500 Index has returned this year while the very lowest quality companies have returned almost 2.5 times the index. As other analysts such as Jeremy Grantham and Mark Hulbert have pointed out, this relative difference in performance of the two groups is off the charts when you compare it to similar market rebounds in the past.

There may be a shift taking place. Since the beginning of the third-quarter earnings reporting season the average high quality stock has returned 7.5 percent, versus 6.4 percent and 5.9 percent, for mid-quality and lower-quality stocks, respectively. Now, that's just a blip on the relative performance graph compared to the extent that low quality has outperformed this year. But it's worth noting because if investors are being more sensitive to actual company operating performance, then the trend in relative performance between high and low quality companies may continue. The graph below shows the rate at which companies beat their third-quarter earnings estimates based on S&P's quality rating. Nearly 90 percent of higher quality companies beat their earnings estimates, while only about 60 percent of low quality companies beat their earnings estimates. If economic and earnings growth over the next few quarters turns out to be lumpier than is currently forecasted, then high quality companies will likely be in a better position to produce more favorable results versus expectations. If investors continue to show discretion in choosing companies with favorable operating performance relative to expectations, this would benefit higher quality companies.

Beat Rates

About 80 percent of the companies reporting earnings in the third quarter beat analyst's estimates. That's a record, according to Bloomberg's record keeping of positive earnings surprises, which began in 1993. We'll use this data to attempt to answer a couple of important questions. Is there value in tracking the earnings surprise rate? Can investors expect the market to move higher when the surprise rate rises? Is there information in a declining earnings surprise rate?

The graph below shows the Positive Earnings Surprise Rate for the S&P 500 Index going back to 1993, using monthly data. As you can see, there has been a secular rise in the surprise rate. Over time companies have caught on that investors prefer positive surprises to negative surprises. That may be part of the reason for "one penny" earnings beats, which are relatively common. The tendency for analysts to lower earnings expectations as the actual reporting quarter approaches is another contributor to the general secular rise in the beat rate. If these practices continue, at some point we'll need to drop the word 'surprise' from the earnings surprise moniker.

But until then, and considering that the earnings surprise beat rate has become so widely followed - and often used as bullish argument - it's worth looking at whether it's a valuable indicator. Often the clearest way to see if an indicator is useful is to test it as if it were a trading strategy. When this is done with the earnings surprise line you get very mixed results. As you can see by the graph above, absolute levels don't work since the index as drifted up over the 15-year period. Information gleaned from simple look-back strategies is also mostly unhelpful. The graph below plots the return of a strategy where you invest in the market when the surprise line rises from the previous month, while earning the risk-free rate during periods where the surprise line fell from the prior period.

It's clear that the level and short term increases and decreases in the surprise rate aren't particularly useful. This is partly because the beat rate data is fairly volatile from month to month.

Even so, we can ask a slightly different question. Given that the earnings beat rate seems to have moved up gradually over time, what happens when - despite this trend - companies fall short of the recent trend, either by a significant amount or over an extended period of time? We can pick up either pronounced or sustained disappointments in the beat rate by taking a 12-month moving average of the beat rate and comparing it to the 24-month moving average. The red line in the chart below shows the investment return from avoiding periods when that 12-month average fell short of the 24-month average.

Since the beat rate has generally trended up over time, the 12-month average has generally been above the 24-month average, so this version of the indicator has the tendency to stay invested. As you can see, the earnings surprise rate was above its moving average during a portion of the late 1990's rally and the bulk of the 2003 to 2007 rally. But what is most impressive is how much of the market's losses coincided with the surprise rate falling below its moving average, which happened convincingly in 2001 and again in 2007. By avoiding market risk when earnings surprises fell below their 24-month moving average, the maximum drawdown using this simple metric was only about half of what a buy-and-hold strategy experienced.

So, the surprise rate is mostly ineffective when measured on an absolute level and also when used with shorter-term comparisons. But there may be some information in significant or sustained failures of the beat rate to keep up with its recent history (e.g. its 24-month average). With the current beat rate at a record high, the current rate may be difficult to sustain in the months ahead (based on typical past trends). Investors may want to watch the rate of that descent.

Here are a few important caveats. While disappointments in the beat rate are important, this indicator leaves out too much information to be used as the sole determinant of an investment position. There's no valuation component, no market action or price trend measures, data availability is mostly limited to two bear markets induced by dramatic profit recessions, and it wasn't tested in a way that an investment model should be tested (split samples with a holdout data set). While the indicator turned down before the major losses of the most recent bear market, it didn't turn negative in the 2000-2002 bear market until about half of the losses were already sustained.

Even so, since the surprise rate is an indicator that has gained in popularity, it is worth highlighting its strengths. The primary strength of the indicator, based on available data, is that becoming more defensive when the surprise rate is declining rapidly may be helpful in mitigating risk.

---

The foregoing comments represent the general investment analysis and economic views of the Advisor, and are provided solely for the purpose of information, instruction and discourse.

Prospectuses for the Hussman Strategic Growth Fund, the Hussman Strategic Total Return Fund, the Hussman Strategic International Fund, and the Hussman Strategic Dividend Value Fund, as well as Fund reports and other information, are available by clicking "The Funds" menu button from any page of this website.

Estimates of prospective return and risk for equities, bonds, and other financial markets are forward-looking statements based the analysis and reasonable beliefs of Hussman Strategic Advisors. They are not a guarantee of future performance, and are not indicative of the prospective returns of any of the Hussman Funds. Actual returns may differ substantially from the estimates provided. Estimates of prospective long-term returns for the S&P 500 reflect our standard valuation methodology, focusing on the relationship between current market prices and earnings, dividends and other fundamentals, adjusted for variability over the economic cycle (see for example Investment, Speculation, Valuation, and Tinker Bell, The Likely Range of Market Returns in the Coming Decade and Valuing the S&P 500 Using Forward Operating Earnings ).