Déjà Vu All Over Again?

Since 2012 began, many in the media have opined the year’s start seems similar to the last two—an implicit forecast regarding what’s ahead in the short term. 2010 and 2011 each started off with accelerating economic data and good market returns, only to see subsequent growth-rate deceleration and sizable corrections (albeit, with slightly different timing). Recently, data like the US Q1 GDP report (which showed decelerating headline growth) and Friday’s US Employment Situation report (showing continued hiring, but at a slower clip than earlier this year) have some claiming the redux is here. For example, Friday’s lower-than-expected 115,000 US nonfarm hires triggered headlines like this. But few seem to stop and ask: Are decelerating economic growth rates reliably predictive of corrections? When framed by a broader backdrop, the answer appears to be clear: No.

First, there’s a bit of a logic problem here. Stocks tend to lead economic activity, not lag it. Economic data, however, lag by definition, as the approximations are published after the economy generated the results. This after-the-fact feature is even more salient with hiring and unemployment numbers—they tend to be very late-lagging indicators, historically. (Which the present bull market illustrates nearly perfectly.) Meaning a market forecast hinging on a jobs figure seems quite backwards.

But even beyond that, there’s very little to suggest slowing economic growth rates (hiring, etc.) are really unusual. All economic data series are volatile. Acceleration and deceleration are commonplace, to be expected and don’t necessarily carry big forward-looking market implications. GDP’s dozens of accelerations and decelerations amid overall expansions and bull markets in the post-war era illustrate this. Simply, there are not nearly enough identifiable market corrections to validate the deceleration-begets-correction thesis.

Some might argue stocks are moving ahead of data as deceleration in some recent data is a sign of what’s to come. But attempting to forecast slower growth rates ahead is a speculative bet if it’s based solely on what’s seen now. In fact, this narrative doesn’t even match 2011’s actual results. GDP decelerated quarter over quarter once last year—Q1.ⁱ The correction didn’t begin in earnest until the summer, while GDP was amid a string of what turned out to be three consecutive quarters of acceleration. Similarly, hiring didn’t decelerate the entirety of the correction.ⁱⁱ But that wasn’t the perception at the time.

And that’s what corrections are mostly about—perception, fear, emotion—sentiment. What drives these perceptions could be anything from slowing growth to fears over Y2K. It could be fears of hot growth sparking inflation leading to a Fed hike. It doesn’t matter how closely fear resembles reality—or anything reality is likely to become. In this regard, a correction is literally always possible as sentiment can change on a dime with little or no warning. Exactly why attempting to forecast a correction is so difficult—there’s just no reliably predictive way to truly gauge the probability an event or fear will quickly turn sentiment enough to affect stocks. And when you consider the fact corrections have occurred in what turned out to be many strong bull market years (1997, 1998, 2006 and more), it really seems quite futile to try.

That corrections are associated with sentiment and not decelerating data is further supported by the selection of data points chosen in support of the argument. In the present, is it really true headline GDP growth deceleration is so predictive? What about accelerating components, like consumer spending? And with hiring, why look solely at monthly reports and not consider the upward trend of revisions after the fact, last week’s sharp dip in jobless claims or the potential data could be skewed by seasonal adjustments? (It’s further telling few folks seem to question decelerating data, but are more than willing to consider better-than-expected data farcical.)

So what does this non-selection of data disagreeing with the slowing-growth-will-cause-correction thesis show? It seems to us those now forecasting a correction aren’t really basing their view on slowing economic data. Rather, it seems more like a behavioral bias in which the brain sees a pattern where one isn’t, extrapolates it forward and tries to justify it. Simply, stocks may do a litany of things in the short term for a whole host of reasons—and a correction is one possibility. But whatever stocks wind up doing in the near future, it won’t be because of—or in reaction to—what happened the last two years.

ⁱSource: US Bureau of Economic Analysis.

ⁱⁱSource: US Bureau of Labor Statistics