Sell in May and go away? Not this year

This was not a good year to go away in May. If you followed that old adage, you've missed a 7.0% increase in the S&P 500 since April 30. Of course, if you knew to get out on May 21 and get back in on June 24, you would have dodged a 5.8% mini-correction and have a gain of 8.9% since the low. But that isn't much better than the don't-go-away return of 7.0%. Previously, I've frequently argued that the trick to this bull market is to keep riding it. That's especially so if this aging bull continues to rage for another four years, which is my working hypothesis--until further notice. The month of May does not stand out as a particularly peakish one. Moreover, during the previous three corrections of the current bull market, it would have been prescient to jump off the bull in April, just before May 1, but you had to know to get back on the bull on July 2, 2010, October 3, 2011, and June 1, 2012. By the way, there was only one correction during the previous bull market. In terms of the average returns of the S&P 500 for each of the 12 months since 1928, July tends to be the best, with a gain of 1.6%. During the 86 Julys since 1928, it was up during 49 of them with an average gain of 5.2%, the best such result of all the months. This year, the S&P 500 slightly underperformed the July average with a still-solid gain of 4.9%. August tends to be a good month too, with 50 (of 85) months on the upside averaging 4.0%. If this August is up by the same amount, then the S&P 500 will be at 1753 by the end of this month. So it's looking like blue skies for the Blue Angels. We calculate these hypothetical paths for the S&P 500 by multiplying forward earnings by P/Es of 10.0-15.0 in increments of 1.0. Based on the latest available forward earnings of $118.26 per share for the week of July 25, the P/E rose to 14.5 at the end of last week. If it rises to 15.0 by the end of this month, the S&P 500 will rise to 1774, or higher if forward earnings continues to ascend to new highs as it has for the past 13 consecutive weeks. Of course, there's nothing special about "15.0" other than that most investors probably view it as the fair value for the P/E. Previously, I've noted that the Rule of 20 would put the P/E at 17 based on a 3% bond yield and 19 based on a 1% inflation rate. Nevertheless, under the circumstances, I would view 17-19 multiples as indicative of an Irrational Exuberance scenario, to which I continue to assign a subjective probability of 30% for this year. Multiples of 15-16 would be consistent with my Rational Exuberance scenario for this year, with a 60% subjective probability. The data suggest that summer rallies are a surer bet than going away in May and coming back in the fall. Then again, the fall season has a bad reputation, and rightly so since it includes September, the worst month for losses. The average return for all Septembers since 1928 is -1.1%. During the 46 down months, September's average loss was -4.8%, the worst of them all. Some of the biggest corrections and bear markets since 1928 started in the fall. Last week, I reviewed some of the challenges that could trip up the bull this coming fall. On September 18, the FOMC will meet and probably start tapering QE. On October 1, the second round of the sequester will kick in. At the same time, Obamacare is scheduled to be implemented. There are mounting risks of congressional gridlock over spending appropriations for fiscal 2014, which start October 1. At about the same time, there could be another impasse over raising the debt ceiling. There is already chatter about a government shutdown. So go away to the beach in August, but stay fully invested. When you come back in September, you'll be rested and ready for the stress tests over the remainder of the year. U.S. Economy. In addition to analyzing the monthly seasonality of the stock market, we have noticed that the first trading day of the month tends to be an especially good one during bull markets. We've attributed this to the release on those days of the manufacturing PMI, which tends to be bullish when the economy is expanding. It happened again last Thursday, August 1. The M-PMI was remarkably strong, jumping from 50.9 in June to 55.4 in July. Its major components all surged as well, with new orders rising from 51.9 to 58.3, production from 53.4 to 65.0, and employment from 48.7 to 54.4. These data certainly support our “Second Recovery” scenario with economic growth led by housing and autos. They also support the view expressed in Wednesday's FOMC statement that while economic growth was “modest” (down from “moderate”) during the first half of the year, it should pick up during the second half. Last Thursday's drop in initial unemployment claims to 326,000, the lowest since January 2008, also supports the more upbeat outlook for the economy. The four-week average remains at a cyclical low and is a good leading indicator of the unemployment rate, which fell to 7.4% in July. That's the good news. The bad news is that Friday's employment report was among the weakest of the year. Let's review: (1) Lots of weak numbers. Payrolls rose 162,000 during July, below the 192,400 average from January-July. The previous two months were revised downwards. More significantly, aggregate hours worked fell 0.1%, and average hourly earnings also declined 0.1%, suggesting that lots of the new jobs are part-time ones with low wages. (2) Not enough full-time jobs. July's household employment survey shows that part-time jobs rose 174,000, while full-time ones increased 92,000. Since the start of the year, part-time employment is up 731,000, while full-time has risen 222,000. The data show that since the record-high peak in employment during November 2007, full-time jobs are still down 5.8 million, while part-time jobs are up 3.5 million to a record high. The biggest losers of full-time jobs since November 2007 have been 25- to 54-year olds, with a loss of 7.7 million. Over this same period, this group has gained only 864,000 part-time jobs. (3) Mixed readings on wages and salaries. Our YRI Earned Income Proxy--which is the product of aggregate hours worked and average hourly earnings in the private sector--fell 0.2% during July. That's bad news for July's wages and salaries in personal income, since it is highly correlated with this series for the private sector. On the other hand, inflation-adjusted wages and salaries in personal income managed to grow at a 3.4% (saar) pace during the second quarter. (4) Robots are coming. Despite the big jump in July's M-PMI Employment Index, manufacturing payrolls rose only 6,000 during July. They are up only 24,000 since the start of the year. They actually peaked at a record high of 17.6 million during April 1998 and are now down to 12.0 million, only 9% of total payroll employment! Manufacturing jobs tend to be full-time and pay high wages and benefits. Some of them were “offshored.” Many have been automated with robots. Ed Yardeni, president and chief investment strategist at Yardeni Research Inc.