Weighting the Employment Reports

James Hamilton wasn’t happy with the report that employment per the payroll survey showed weak job growth. Dr. Hamilton notes the negative employment growth per the household survey plus another way of measuring employment growth:

The ADP estimate is based on actual payroll data, not a survey, and covers twice as many individual businesses as the BLS. When you add in the public employees (64 + 25 = 89), ADP called the nonfarm payroll number essentially spot on this month two days before we heard from BLS.

Earlier this year, he advocated combining forecasts and yesterday he gives an example:

For April that calculation comes to

(0.8) x (88) + (0.1) x (89) + (0.1) x (-468) = 32,500 new jobs.

With 25,000 of those jobs coming from the government, I’m thinking we could have had essentially zero net job creation from the private sector in April.

If one changes the weight for the payroll survey to 60% with 20% weights for the other two series, overall employment growth would have been estimated to be negative 23,000. Not that one would have any reason to prefer one weighting over another – but notice when the household survey and the other two series give such divergent signals, the weighting matters. At least, we don’t have Treas. Sec. Snow’s method – report whichever series gives the highest employment growth figure and declare that series to be the most reliable for at least the current month.