We've covered before how the Federal govt uses two surveys of employment; the establishment and household surveys - and although each surveys different segments, each usually report the same levels of employment, even if they occasionally diverge during, ahem, presidential election years.
But here they have diverged widely since March: the Establishment survey shows 2.7 million new jobs, while the Household survey shows 12,000. I'm a firm believer in Hanlon's Razor, so I don't believe it's a conspiracy. Two things instead seem to be happening with the data:
- New business starts are up and the BLS is overweighting this impact on new job growth. It is well known (even by the Fed) that new business starts rise during a recession as people who are laid off start new companies, often with severance packages from their former employers; and
- we are seeing employment fall in while average hourly wages are increasing - which could be explained by layoffs + severance packages. This is exactly how severance packages would show up in the data.