Union Membership--Good News? Or Not?

Both Brad deLong and Matt Ysglesias have posts referring to the rise in union membership from 12% of the labor force (in 2006) to 12.1% (in 2007). (Brad's post is, essentially, a link to Matt's).

I'm getting around to this somewhat late, but let's not get all giddy here. The BLS data indicate that union membership in the private sector fell from 7.5% to 7.4% between 2006 and 2007. And that it also fell in the public sector, from 36.2% to 35.9%. The only way for this to lead to an increase in the overall rate of unionization--from 12% to 12.1%--is for the public sector's share of total employment to have increased. (Even looking at the percentage of workers represented by unions--not all of whom are union members--that also fell in the private sector, from 8.2% to 8.1%, and in the public sector, from 40.1% to 39.8%.)

So declining union membership both in the public and in the private sector becomes a story about rising union density. Both god and the devil are in the details, and the details do not support the story.