Friday's Job Report Was More About Bad Data than Bad Jobs


Dealing With ‘Garbage Data’

The old computer adage 'garbage in, garbage out' seems to be relevant with this week's employment noise.

ADP had a big gain (+250k vs expectations of +190k) and had a modest downward revision to last month.

The NFP establishment survey missed (148k vs the same 190k expected) but had upward revisions.

The household report (which is even more volatile and less consistent) was had smaller job gains.

Average hourly earnings ticked up (though with a downward revision to last month) indicating the headline miss may have been misleading.

The average number of hours worked during the week stayed constant, another indication of strength relative to the weaker headline number.

Finally, the notorious Birth/Death model which attempts to model jobs at newly created companies versus those that closed had losses of 38k jobs. Given how favorable the new tax policy is towards small business, I think the possibility that number is wrong and overstates job losses, is high.

This is a real issue for companies trying to plan and forecast their needs.

In a world where retailers know what I searched for before breakfast and can fill my e-mail and websites with targeting information, it is sad that we have such poor quality data for something as important as the number of jobs the largest economy in the world is creating.

I would like to see the government invest in data collection as it should help the entire business community.

From the markets side, the weak headline number will largely be ignored, as it should be, so expect more of the same - some pressure on bond yields, but not enough pressure to derail stocks - yet.