Last Friday, as it does the first Friday of every month, the Department of Labor’s Bureau of Labor Statistics (BLS) released the monthly report on employment in the United States. The widely-followed estimate for total payroll employment fell by 701,000 jobs, a massive one-month decline that significantly exceeded analyst expectations.
The monthly employment report includes two main surveys, however – the payroll survey, a survey of firms that led to the result above, and a survey of households, from which the unemployment rate is reported.
The household survey doesn’t just measure unemployment. It also includes an estimate for total employment, like the payroll survey. And historically (for example, amidst the recovery from the 1990-1991 recession) the household survey has provided a more reliable read on labor market conditions.
What did the household measure for total employment do last month? It fell by almost 3 million jobs, a nearly two percent decline in a single month, and a decline about four times as large as the decline in payroll employment.
Granted, the household survey tends to be more volatile than the payroll survey. But it still is worth watching, given the sampling and other issues affecting both BLS surveys.
Amidst the challenges in interpreting government financial results, given their accounting issues, it is also worth remembering that sometimes, we have enough of a challenge just counting heads.