Keep Fading This Rally

Includes: DIA, QQQ, SPY
by: Kevin Flynn, CFA


The spring rally in equities is as perennial as it is beguiling.

Earnings and economic data are telling a different story than the rebound.

Some thoughts on employment.

We live in a world of conflicting signals, don't we? In January and February, stocks were taking a beating. I wrote a series of articles called "The Four Horsemen of the Recession," detailing weakness in growth (or lack thereof) in wholesale sales, industrial production, consumer spending, and business investment. None of these categories caught as much attention as the fall in equity prices.

Later in February, stocks rebounded from the long-term trend line that underlies the current bull market. As I noted in my last article for Seeking Alpha, this was accompanied by a series of forecast errors - not actual data in most cases, but forecast data - that seemed to validate the rebound in stock prices. Corporations rushed in to buy more stock after the end of earnings season, and indeed the modern earnings cliche has become "revenues are down, earnings are down but don't worry we're going to borrow billions and use it to buy back stock."

Now here we are in the midst of the spring rally - a sacred rite on Wall Street, even in the midst of bear markets and recessions - with stock prices sitting at or just above where they started the year. So of course, all is well.

You wouldn't know it by looking at retail sales. The latest report was for March, showing a seasonally adjusted decline of (-0.3%), well below expectations for what would have been at best a quite tepid gain of only 0.1%. February was revised upward to 0.0% and I suspect March might get a tick upwards later too, as the seasonal adjustment process is always made tricky by the shifting of Easter.

The trailing-twelve-month ("TTM") rate for unadjusted retail sales has actually improved with the latest data, from 2% in January to 2.5% through March. However, this is a deceptive improvement, as it benefited from a double-barreled quirk in the calendar. The first is the additional (leap-year) day in February, with a simple interpolation suggesting that the rate would be 2.2% otherwise. Second, the current TTM rate incorporates TWO Easter holidays, the one from April 2015 and the one in March 2016. Easter is one of the three reliable holidays where actual spending can be relied upon to increase. Given this relatively weak performance, the negative initial estimate for March looks more reasonable.

Here's another little data tidbit to make you think - year-over-year retail sales growth for the first quarter of 2016 (unadjusted) currently stands at +3.64%, which looks good at first blush, but is well below average for a leap-year quarter. The only leap-year quarter that was worse in the last twenty years was 2008 (-0.32%), during the onset of the Great Recession.

You wouldn't know all is well either by looking at wholesale sales. The latest reading came last Friday for February, showing an initial estimate of another decline (-0.2%). Although the TTM rate leveled off (unadjusted), thanks to the extra day in February, it still stands at (-3.7%), an alarmingly low negative that has historically been associated with recessions. The latest data on business inventories (which include retail goods) show the inventory-to-sales ratio still at 1.41, the highest level since the recession and before that, 2001 (the previous recession). With inventories too high and sales weak, the inevitable rebalancing will be a negative for output and GDP.

data through February 2016, source US Commerce Dept.

Of course GDP isn't going to tell you that all is well either. The current tracking estimate from the Atlanta Fed is for a 0.3% real rate in the first quarter (seasonally adjusted and annualized), down from 2.7% two months ago. But employment (a lagging indicator) and the stock market rally say otherwise, so the solution is to pooh-pooh the GDP number, from the Federal Reserve to the various Street strategists complaining about the weak first-quarter bias. Well, let's see. So far this quarter, business spending is negative, cap-ex is negative, trade is negative and retail spending (the PCE kind, not the Commerce Department number) is estimated at 1.8%. So where exactly is the hidden strength that the GDP report is missing? Last year and the year before, we had the excuse of two severe winters. This year the excuse is that the data must not be right. Just look at the stock market.

Ah yes, the stock market, where earnings estimates for the first quarter have come down from +0.7% at the end of 2015 to a current (-9.1%), the biggest one-quarter change since 2009 and the recession, according to FactSet. In a nod to the analyst practice of contriving low estimates that can be beaten, FactSet is now estimating the implied expected rate to be between (-5%) and (-6%). That's wrong too, don't you know, because all you have to do is eliminate the sectors that are down year-on-year and you have positive growth. I can't argue with that logic, though what will happen when we run out of positive sectors is unclear.

But the stock market is telling us that all is well, unlike the bond market, where yields are running near three-year lows (ergo, it must be wrong too). Just ask asset managers, well known for telling you that every time is a good time to invest, just as every day is a good day to buy a home in real-estate land. I was surprised to read a recent missive from Pimco that declared that weakness in manufacturing was ending - based on one month's read of the ISM survey - and that the Fed and Janet Yellen were willing to "run the economy a little hot."

The February inventory-to-sales ratio for manufacturing remains at 1.37, the highest it's been since 2009. That doesn't bode well for the weakness in manufacturing. The latest data on inflation show the PPI and CPI both weakening somewhat in March, with year-on-year PPI negative and year-on-year CPI at 0.9%, the ex-food and energy component of the latter declining from 2.3% to 2.2%. No doubt that report is missing something too, but what is running hot besides short-covering?

I also last wrote that you should fade the current rally, and I stand by that advice, despite the 2% gain in the index since that time (most of it in the last two days). Of course stocks could move back to 2100 on the S&P, I've been doing this too long to say that stock prices can't do anything. The "pain trade" is still higher for now, something to keep in mind. What I am saying is that the data - and that includes earnings at the "Gang of Four" banks, all of them down on a comparable basis this quarter, including JP Morgan (NYSE:JPM), which excited everyone yesterday with their down quarter that beat expectations - is telling you otherwise. When stock prices are falling, it's fashionable to talk of recession, when they are rising, it's fashionable to discard such talk as the nonsense of disgruntled perma-bears. But earnings and the data are telling you that the business cycle is ending.

I'll close this note with a reminder on employment, because after stock prices it's the most-often cited statistic that recession cannot possibly be lurking. First, it's a lagging indicator. Second, reaching full employment (where we are now) doesn't mean that incomes will grow forever and thus forestall any downturn in the economy - if things were that simple, we wouldn't know what a recession was. Finally, I noted last year that employment entered its peak period in August, and that historically peak periods had lasted from six to eighteen months, with a median of about a year. We only recently passed six months, so the current data hardly comes as some sort of robust contradiction. Job growth is weakening, if you know how to look, but it's weakening slowly and gracefully, rather nice in face of the endless corporate attempts to boost earnings by shifting jobs overseas.

It just won't stay that way. You may have read or heard on the radio that weekly jobless claims are at their lowest level since 1973, so it may be natural to assume that recession must be far off indeed. Tastier yet is that cake-icing that reminds us that the employed population now is much larger, so claims as a percentage are smaller. Both facts are true, but ignore two less pleasant realities that are not likely to be reported on national reports anytime soon. The first is that relative to the total workforce, our manufacturing employment is only about a third of what it was in 1973, and manufacturing is far more susceptible to layoffs, especially temporary ones. Service sectors typically don't start to layoff until recessions are well underway.

The second is that weekly jobless claims were getting better all throughout 1973, on the track they had been on since the expansion part of that cycle had started, until they didn't, in October 1973. Claims started getting worse in November, and that is considered the begin date of the 1973 recession. Not much of a warning, was it? Beware of using any one data series as an infallible leading indicator - indeed, one of the largest weights in determining when a recession is beginning is a long-lasting downturn in employment, and the long-lasting part isn't certain until long afterwards.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.