We Don't Need More Employment Growth Nowcasts!
Slack Nowcasts Are the Missing Piece
This morning I saw this tweet by the great Joe Weisenthal (you can read the Carlyle report in question here):
My first instinct, of course, was to reply to Joe with my own joke:
But then I added something more substantive: we don’t need any more nowcasts or estimates of employment growth. Even before the government shutdown, employment growth’s value as a cyclical indicator was significantly diminished by the impact of labor supply growth fluctuations. News reports characterized the Carlyle estimate of +17K in September as weak, but we don’t actually know that for sure since we have only a blurry idea of the breakeven rate.
What we do need to track is slack - the kind of stuff that comes out of the currently-unpublished household survey.1 So please, if you’re sitting on a bunch of valuable proprietary data, please consider: we need is more nowcasts or indicators of slack, ideally from hitherto-unused measurement sources.
We don’t have many of these. The Chicago Fed unemployment rate nowcast is one. (It assessed the September unemployment rate at 4.34%.) The LinkedIn data on hiring and separations is another option, whether published by LinkedIn itself or by Revelio.2 And we have Indeed’s wonderful job openings data, and the Conference Board’s job market differential. Morning Consult has their own survey-based estimate of the unemployment rate that should probably get more news coverage (but is limited to subscribers).
If some bajillionaire with money to burn wants to make me happy, please conduct a very-large-sample, free-to-the-public survey inspired by the CPS!
Currently unpublished, and if the shutdown continues much longer, also uncollected!
I wish LinkedIn would publish the history of the data, and that Revelio would publish their data without 12 month smoothing. But beggars can’t be choosers.