Sunday, 8 May 2016

How not to analyse the effects of a minimum wage

From the National Employment Law Project comes this piece of "analysis" of the effects of the minimum wage: Raise Wages, Kill Jobs? Seven Decades of Historical Data Find No Correlation Between Minimum Wage Increases and Employment Levels by Paul K. Sonn and Yannet Lathrop.

At the end of the second paragraph of the Summary comes this analytical and policy gem,
Rather than an academic study that seeks to measure causal effects using techniques such as regression analysis, this report assesses opponents’ claims about raising the minimum wage on their own terms by examining simple indicators and job trends.
So they do not even try to sort out the actual effects, and causation, of having a minimum wage, they just look for a correlation, or lack of a correlation, and this will do.

And yes, stopping reading after the second paragraph is therefore optimal.

6 comments:

Mark Hubbard said...

I was going to comment there is not more butchered debate than over the minimum wage; however, I then remembered the related debate growing around a UBI.

Mark Hubbard said...

... That's no more ....

Tim Worstall said...

That NELP report does actually get gloriously worse. They count the total number in employment over the years. But do not adjust for the growing population.

Thus their argument is that an economy of 340 million people has more jobs in it than one of 160 million people. See! Minimum wage rises don't cost jobs!

what's worse is that I've already had people pointing to that report as proof that the min wage doesn't cost jobs....

Paul Walker said...

Tim. This doesn't surprise me. What does surprise me is that NELP would run with this standard of crap. I'm sure they can get people to do better analysis than this, so why run with a report this bad?

Tim Worstall said...

"so why run with a report this bad?"

It might be necessary to mention that dread word: "politics"

Paul Walker said...

Politics, a dirty word if ever there was one.