Reply #10: Again, payroll ISN'T the "gold standard"

[Frodo]

They are both completely valid surveys for what they measure. They merely contradict each other recently. And this is not a new event. I read a recent study that showed that republican presidents consistently do better by the household measurement than the establishment survey while Democrats establishment survey progress outstrips their household measurement.

Note that it wasn’t comparing Republicans to Democrats, but their relative performance based on how that performance was measured. I’ll see if I can dig up the report.

It would seem that they create different types of jobs. Or more likely, that varying tax policies effect the reletive NEED for jobs (on the margins)

But back to the “gold standard”. What was the employment picture under Reagan? Under Carter? Under Ford? People are FAR more likely to be able to tell you what the unemployment RATE was than the number of jobs there were. Common historical comparisons from year to year are done with the household survey. ECONOMIC evaluations of one month to another are more commonly done through the ESTABLISHMENT survey. That's why Greenspan wants to look at one more than(but not to the exclusion of) the other, he need timely measurements of changes because things fluctuate. But overall multi-year results including all goernmental (not just Fed) policies may be more accurately measured by a survey that take in to account who WANTS a job. Give every family in DC a nice home and you'll see the "unemployment rate" there plummet even though the local version of an establishment survey would show no new jobs.


But to your first point. What "bias" have you identified? This is the same process the census uses to identify households for the more detailed survey.