The labour reports for January 2014 for the U.S. and Canada were good from my point of view. However, most investors apparently look at the Nonfarm Payrolls number (from the U.S. Establishment survey), and disagreed with this assessment.
The Nonfarm Payrolls number is based on surveying employers, and was a sub-par 113,000 in January, following a weak December (and that weakness was not revised away). The initial reaction of the Treasury bond curve was to rally, erasing yesterday's losses.
However, the Household Survey, which is a sample of households (as the name implies) was good. The unemployment rate fell again to 6.6%, but this time for good reasons. The participation rate actually rose, which meant that the employment-population ratio rose 0.2% to 58.8%. (The number of jobs created on this measure is a whopping 616 thousand!) The Household survey is based on a smaller sample, and is noisy from month-to-month (which presumably explains why the market does not key off of it; investors hate "whipsaws"). However, the January numbers are consistent with the glacial improvement in the labour market that we have seen over the past few years.
The Canadian Labour Force Survey numbers for January were also good, with the unemployment rate dropping by 0.2% back to 7.0%. The Canadian equivalent of the Establishment Survey is produced with a lag (they just released the November data), so most people work with the noisy survey of households. This report seems to indicate that the disaster in December was probably noise. Once again, mediocrity reigns.
I want to do a review of labour market trends "soon", so I will just leave you with a link to Calculated Risk for those who want to get their chart fix.
Have a good weekend.
(c) Brian Romanchuk 2014