Recent Updates

  • US: BEA Auto Sales, Employment Situation (Nov)
  • US: Household Survey Detail (Nov)
  • Canada: Labor Force Survey (Nov)
  • US: Establishment Survey Detail (Nov)
  • Romania: Foreign Reserves Press (Nov)
  • more updates...

Economy in Brief

Payroll Job Data Not A "Fluke": Reflects Slower Growth in Q1
by Joseph G. Carson ([email protected])  March 11, 2019

Payroll employment increased by only 20,000 in February, following a 311,000 gain in January. Many analysts quickly dismissed the small gain in payrolls, arguing that it was inconsistent with other labor market indicators and some went as far as calling it a "fluke". Yet, if there was a "fluky" number it was January’s payroll gain and February’s figures were merely a correction to a temporary and inflated gain of the prior month, bringing job growth more in line with current estimates of Q1 GDP growth.

The payroll data published by the Bureau of Labor Statistics (BLS) is based on the number of employees who worked for private and public establishments and received pay for any part of the survey pay period. In this survey, people could be counted more than once, if more than one employer pays them. In January, something very unusual occurred in that approximately 800,000 federal workers were not at work due to the government shutdown, but were still classified as being employed since they would eventually be paid for their time away from work.

How many of these employed/not at work federal workers went looking for temporary work until the government shutdown was over is hard to say. Yet, it is possible to glean some insight on this issue by inspecting job data from the household survey.

The household survey records the number of people working, and it also classifies people working/moving from full and part-time jobs. In January, the household employment report showed an unusual jump of 602,000 people working part-time for "slack work or business conditions" in nonagricultural industries, followed by equally large decline of 655,000 in the same category in February.

The volatility in these numbers is very unusual even for a job category that is known to be notorious for being volatile. Nonetheless, the data does offer some proof that there was some unusual movement in the workforce in the last two months that could be attributable to a number of government workers being counted twice in January’s payroll data as they gained temporary part time work and only one time in the February report when they left their part-time job and returned to their regular full time jobs in the federal government.

Due to conceptual and measurement difference one cannot apply the household data to the payroll series, but judging from some of the business establishments that employ a large number of part-time workers---leisure, retail and transportation---these firms added 120,000 jobs in January and reduced headcount by nearly 20,000 in February.

Even if that sign was reversed and then doubled it would still leave the payroll job gain in February at a relatively low number. Yet that should not come as a surprise. Consensus estimates of Q1 GDP growth range from 0.5% to 1.5%, so its not surprising to see relatively slow job creation---it’s a barometer on how slow the economy is growing in Q1, and should not be characterized as a "fluke."

Viewpoint commentaries are the opinions of the author and do not reflect the views of Haver Analytics.
large image