He worked there 39 years, now has no job and no pension.
The pensions are probably ok. PBGC (Pension Benefits Guarantee Corporation) takes over pension payments for such situations. I get a small pension from a company that went belly up many years ago. PBGC was set up for exactly these situations. He should contact them now, or, wait, and they'll probably write to him about it. Pensions, at least mine, don't have inflation clauses, but, it's still something.
the Japanese cleaned our clocks in the automobile industry
True, they did, but if you're talking about last century, I think there's a little more to it. In a statistics class ('81) the prof talked about this. Demming (you can wiki him) was a sort of genius or at least a trailblazer in applying statistics to quality control. His involvement with Japan goes way back. They applied his methods, while in the US, these methods were totally available and offered, but were ignored and rejected in favor of just turning out as much product as fast as possible. Good QC means you test a certain percentage of product and statistical methods tell you how many you should test to be representative of the total, and what the bad rate of product is and where the common problems are. It's nothing new. Probably part of the issue is, it costs to do it right, so no doubt a lot of decisions are made (poorly) by asking, what's it cost for good QC, compared to what's the chance something will require a recall/lawsuits and how much does that cost. It's shortsighted. You can read about his philosophy in the wiki article. "...the bulk of the causes of low quality and low productivity belong to the system and thus lie beyond the power of the work force". It doesn't help to just scream "zero defects" and "more productivity" at the workers, it just makes things worse.
If you build a million vehicles, you don't test a million. You test some number that is high enough that the chance you are missing a major problem is close to zero. That's not the same as zero defects, in manufacturing there are always defects. Now something like artillery shells, at least in the old films, they probably test them all, because if the shell is oversize it can result in catastrophic failure/deaths. For vehicles, if you pull x number off the line at completion and test them, you'll find out pretty quickly where the problems are and statistics tells you how many to test to achieve what level of confidence. You will not reach 100% good. But that costs money to do, so if you have the philosophy to build one trans, and make it work in all your vehicles, realizing yes there probably will be a bunch of problems, but, how much will that cost compared to building 5 different transmissions - well - you see the problem. "The perfect is the enemy of the good." You can target all perfect, that's fine, but you have to accept and realistically target something less (97%? 99%?) or you never get anything done.
I think more and more of vehicle manufacturing is robotic. That doesn't eliminate defects though, and you still need people to run and maintain the robots.
Like only working for a couple hours and finding a place to sleep for the rest of the shift.
Yup this is a problem. Funny, just yesterday I was thinking about working in a paper mill in the 70's. I was a young guy working there summers. I thought, if you work, get paid, you are supposed to be doing something. I kept asking the foreman what am I supposed to be doing? Finally he said "find a dark corner somewhere and take a nap." So there's plenty of blame to go around, partly at least based in no pride in workmanship and the adversarial relationship "salaried" vs "union". It's no wonder employee costs can be nuts. I remember if a machine went down, first you have to call an electrician for the lockouts. That maybe takes 5 minutes but he had to be called in, so, 4 hours probably. Pipe problem? need plumber. A bolt? Need a millwright. On and on. A lot of guys made a real good living taking the Sundays and holidays and overtime. You could get triple time (holiday pay plus Sunday) plus 1.5 overtime of all that, do the math, it's a lot.
I remember when the rail car company went belly up here, a while back, guys poured nuts and bolts down all the big drains and laughed. It's just representative of the adversarial relationship labor/management. You have workers trying to make a buck, don't really care about the product. There's sabotage to 'get back' at management. You have stockholders asking for return on investment, top management taking obscene bonuses and salaries, and the consumer is caught in between all of it.
Speaking of stock (Blmpkn), you could have made a killing on Ford - it was close to 11 and then went to 15, so if your timing was good, you could have made almost 50% on your money in a matter of months, maybe weeks. I was waiting for it to dip below 11, or even 10, it never did. Plus afaik they pay dividends pretty regularly, and, their price/earnings ratio is low. So all in all, there are lots of worse investments. I put money in T-bills, gets amost 5.5%, no risk, but here's the problem. That's barely keeping pace with inflation, so while it's nice to think, I made x on it, actually in terms of buying power, the money is worth no more, probably less, after the returns. The only nice thing about it is I'm not likely to wake up one morning and find out "US Treasury declares bankruptcy" because at that point everything worldwide would collapse.
I like to look at it and tell myself, great, I already made enough on T-bills to pay for new tires. The problem is, if I spend that, it's not a "freebie" because even though I have the same number of dollars I started with (after buying tires), those dollars don't have the buying power they did in the past.
Welcome Visitor! Please take a few seconds and 
