Friday, April 17, 2020

Rules of thumb

FRED gives the unemployment level -- the number of people unemployed -- in thousands of people. I divide that number by 1000 to get the number of millions unemployed.

Then, because it might be interesting, I divide the unemployment rate by the number of millions unemployed:

Graph #1: Unemployment Rate per Million Out of Work
Starts around 1.6% unemployment for each million people out of work. Ends around 0.6%. Population is bigger now, that explains it. Also, the percent of the population that's in the labor force. And how they count unemployment, too: Probably not a big effect (on this graph), but that's just a guess.

Here's something odd: Take the first graph, put a second line on it, and have the second line show the inverse: Turn the ratio upside down.

Graph #2: Rate-to-Level (blue) and Level-to-Rate
The one line starts at 1.6 and ends at 0.6; the other starts at 0.6 and ends at 1.6. Pure coincidence.

Anyway, these days, throw a million people out of work and it adds about 0.6 to the unemployment rate.

These days, for each 1% the unemployment rate goes up, figure it means an additional 1.6 million people are out of work. So when the unemployment rate increased by 0.9, from 3.5% to 4.4%,  it meant that 0.9 times 1.6 million people -- about 1.44 million people -- had lost their jobs.

1 comment:

The Arthurian said...

Yahoo Finance, 17 April 2020:
"At least 22 million people have filed for unemployment benefits in the last four weeks."

Okay. If my ballpark calculation is right -- though I know it is very rough -- throwing a million people out of work adds about 0.6 to the unemployment rate.

Throwing 22 million people out of work adds about 13.2% to the unemployment rate.

If we're starting at 3.5% and add 13.2, that's 16.7% unemployment. If we're starting at 4.4% and add 13.2, that's 17.6% unemployment.

February = 3.5%
March = 4.4%
April = ?