Saturday, July 29, 2017

Recession? Neh


Two measures of employment:

Graph #1: Two Measures of Employment (Percent Change, Monthly)
The red one is Copyright, 2016, Automatic Data Processing, Inc. ("ADP")
If it is reasonable to expect a premonition of recession in the form of a year-long (or longer) decline toward zero, then it is reasonable to say that this graph does not see a recession starting in the next year or two.

Argument the other way? Off the top of my head, if you want to pull teeth, you can see red rise above blue in 2006, just at the start of a two-year decline to zero. And you can see red again rise above blue in recent months (Nov 2016 - March 2017). That could be telling us something. But there is no red data from before 2002 to show anything similar.

Or you could see increase from 2012 to a high point in early 2014, and "decline toward zero" since then. But the rate of decline (if it is a decline) is so slow it might take 20 years to reach zero.

One nice thing about this indicator is that it is monthly.

See also My recession indicator says "not yet" from six months back.

2 comments:

The Arthurian said...

For an unrelated indicator of recession see

An Indicator of Peril
https://www.720global.com/article/an-indicator-of-peril-598b0231d5bc1
Gross Value Added (GVA) and Real Value Added (RVA)

The Arthurian said...


In Term Spread, September 5, 2017 Menzie Chinn looks at the spread between 10-year and 3-month Treasury rates:

"Run the probit regression... the implied probability of recession is 10% for February 2018."

Low probability of recession.