ZeroHedge and EJ Antoni assert sure, counting on the family sequence…
Determine 1: Whole change in civilian employment (daring black), change attributable to authorities employment (blue bar), to personal (tan bar). Supply: BLS CPS, and writer’s calculations.
If it appears implausible that authorities employment ought to flip-flop between optimistic and detrimental values, then you definately is likely to be excused. The family survey is beneficial for a lot of issues, however determining month-to-month modifications in subcategories is just not essentially one in every of them. As an illustration, evaluate in opposition to the corresponding graph, utilizing the institution survey knowledge.
Determine 2: Whole change in civilian employment (daring black), change attributable to authorities employment (blue bar), to personal (tan bar). Supply: BLS CES, and writer’s calculations.
The family survey is well-acknowledged to be extremely unstable, as proven in Determine 3:
Determine 3: First log distinction of CES nonfarm payroll employment (blue), of CPS employment adjusted to nonfarm payroll idea (crimson). NBER outlined peak-to-trough recession dates shaded grey. Supply: BLS, NBER, and writer’s calculations.
So too for presidency employment (in % phrases):
Determine 4: First log distinction of CES authorities employment (blue), of CPS authorities employment (crimson). NBER outlined peak-to-trough recession dates shaded grey. Supply: BLS, NBER, and writer’s calculations.
Because of this ZeroHedge was pressured to resort to utilizing CPS knowledge, and CES not-seasonally adjusted numbers. Wow.