- XE Contributor
The first bit of soft data is motor vehicle sales, at a 12 month low in March (h/t Calculated Risk):
Motor vehicle sales tend to plateau for long periods of time during expansions, and decline from that plateau roughly half a year before the onset of recession. While there is no doubt that March was soft, I would have a to see a number under 16 million annualized for me to be concerned that this was anything other than a one-month variation.
The next bit of light data comes from PMI Markit, which has launched indexes to compete with the venerable ISM manufacturing index. The breathless news is that the index decelerated month over month:
and the new orders index was at its lowest point since October 2016. The problem here being, well, here's the direct quote from PMI, which didn't publish the exact new orders reading:
"New orders rise at weakest pace since October 2016"
Oh. New orders actually *rose.* They simply rose at a decelerating pace compared with the last few months.
Meanwhile here is the number from ISM, which has a pretty good 70 year record of forecasting recessions whenever the reading is less than 48:
and new orders? Still rose at a pretty good clip:
I would be remiss from completeness if I didn't mention that the S&P 500 hasn't made a new high in a month:
Down all of 2%.
And the 4 week average of initial jobless claims similarly hasn't made a new low since the end of February:
Initial claims would have to rise to at least 280,000 to make me concerned.
This is one reason to separately review short leading indicators and long leading indicators. If the short leading indicators have taken a breather over the last month, residential construction was also released Monday, at that made a post-recession high:
Which means that both of the least volatile measures of the long leading housing market (the other being single family permits, blue in the graph above) are in good shape.
And there's nothing going on with interest rates at the moment that is worse than neutral, although long term treasuries and mortgage rates have fluctuated with mild negative readings in the last several months. Here's the yield spread for both the 10 year minus 2 year treasuries (blue) and 10 year minus 5 year (red):
The latter is much less affected by the Fed's long-time zero interest rate policy than the former, and for the last 60 years both have inverted before a recession. While the 10 year minus 5 year is slightly below the +0.50% spread which would make it a neutral, the 5 year minus 2 year is well above +0.50%, which means it is still quite wide and consistent with expansion.
I'm simply not going to be concerned about a pause in some of the short leading indicators, where the long leading indicators have not turned negative in any consistent way.
Sent from my iPad