The Expert Who Cried Recession

Lots of people have been talking about the next recession and it’s not even here yet (as far as we know). Most have been trying to read the tea leaves and project the year of commencement of the next recession…the second half of 2019, 2020, 2021, etc… In this day and age with all the information and the tools we have (as well as all of the “experts” paid to offer their opinions), it’s natural to be thinking about the next recession after such a long time has passed since the previous recession.

It’s evident now global economic growth is slowing, especially in important markets like China and Germany. While 2018 could be characterized as the year of the tightening (monetary policy), 2019 may end up being a partial reversion of sorts to the looser monetary policy that has dominated much of the past decade. That can’t be good a good sign for economic growth. I suppose central bankers may chalk it up to the engineering of the oft-mentioned yet elusive soft landing. Regardless, we’re observing some softening signals in the U.S. but not necessarily anything to indicate a recession is at the doorstep. Sure, the bears will likely grab hold of February’s weak job report but it was on the heels of a very robust January report. Average the two out and the job growth picture is still in fairly good shape, even if it is a lagging indicator.

One of the more useful economic cycle indicators is the four week moving average of initial unemployment claims, which hit its trough this cycle (for now) in mid-September last year. A rise in initial unemployment claims has accompanied every recession for which there is data. Most often, the increase in claims begins modestly before the onset of a recession and ramps up dramatically from there as depicted in the chart below. You can also see we’ve started a short-term ascent in claims which may be indicative of an oncoming recession or it may be a reflection of various factors including companies responding to tougher business conditions resulting from the trade-war between the U.S and China or maybe even the tightening in monetary policy. Of course responses to either of these could end up being self-fulfilling prophecies of sorts as human behavior/response to perceived reality often times creates its own reality if the herd is large enough.

The Stock Market

Historically, the stock market begins transitioning to a downtrend at about the same time initial unemployment claims begin their transition into an uptrend. For this reason, many strategists and investors keep a close eye on the trend in unemployment claims.

Interestingly, the stock market reached its most recent peak for this cycle about a week after the four week moving average of initial unemployment claims hit its most recent nadir for this cycle in the middle of September last year.

Whenever the initial claims moving average has begun to increase in advance of and into a recession, the stock market has always meaningfully corrected in response, for as long as we have data, as can be seen in the chart below. This wasn’t apparent in the long-term trend-line above but by looking at year-over-year percentage change we can see how jittery the stock market has become historically when claims transition into an uptrend.

What’s interesting about this time around is we haven’t observed a situation historically where claims started to rise, the stock market sold off and then bounced back as strongly and quickly as it has this time. This goes without saying but the sell-off in 4Q18 was most likely primarily driven by rate hike fears and, maybe to a lesser extent, trade wars. As both of those issues have appeared to begin to move in a direction investors favor and broader economic fundamentals have remained relatively healthy, the stock market has bounced back nicely even though the initial claims moving average has continued to tick up.

Imagine the coming volatility if the initial claims moving average continues to trend upwards and if the direction of either rate hike expectations or trade war resolutions reverses. Brace yourselves.

Note: The four-week moving average of initial unemployment claims can be rising or in a rising trend for some time before the actual onset of a recession. The historical average number of months from the initial claim moving average trough to the beginning of a recession is a little over 13 months. Historically, it has been as short as four months and as long as 22 months.

Source: BEA