I've been taking welding classes at my local community college (Washtenaw CC, just down the road from University of Michigan) around my day job as a controls en

I've been taking welding classes at my local community college (Washtenaw CC, just down the road from University of Michigan) around my day job as a controls engineer in the auto industry. This program produces people who win national welding competitions on a regular basis, but they're very open that you don't have to complete the entire thing. The instructor for my summer blueprint welding class kept talking about open jobs in the region and how to get into it -- what certs you needed and what skills to practice. Crappy jobs that didn't require certs paid $20/hr starting + benefits. A couple of years in and with several certs under your belt in a union -- pipefitters or ironworkers -- you could be making north of $40/hr. That doesn't sound high-paying, but many of the people at that level are nomadic, coming into a plant being built and working 12 hrs/day 6 or 7 days a week. Which brings me to my main point -- the statistics on this are heavily skewed. The guy that does that nomadic thing spends some number of months on until he quits and takes 3 months off. Other guys with no certs make money welding art or fixing simple things (a much, much larger category). The BLS numbers aren't a question of what you can make, they're a question of what the distribution looks like.