4 ms·
I don't disagree with the article, but the false equivalency at the beginning is pretty annoying. >It should be obvious that this story doesn't work: Boeing wo
by 34679 7y ago
I don't disagree with the article, but the false equivalency at the beginning is pretty annoying.
>It should be obvious that this story doesn't work: Boeing would most certainly be in trouble if they had to pay their workers sufficient to afford a new jetliner.
You could go the other extreme and ask the same question about paperclips, but paperclips are a poor substitute for automobiles. The equivalent of a jetliner in Ford's day would probably be closer to the Titanic than a Model T.
- mjw1007 7y agoRight. I think Worstall is correct that the story only makes sense if the pay decision can be generalised to other companies, otherwise it won't have enough effect on the market. But there's no good reason to generalise "Ford workers should be able to afford cars" to "company X workers should be able to afford company X products". Rather it should turn into something like "Factory workers should be able to afford cars".
- SilasX 7y agoNo, that's the point[1]: it's dubious to think that the worker's wages should bear any relationship to the finished product.[2] That's all the author is establishing there. Yes, paying workers just enough so they can afford a paperclip is obviously a bad idea. Yes, paying workers just enough so they can afford a 747 is obviously a bad idea. Yes, it's still a bad idea when an efficient wage would be just enough for them to afford the same product (here, a Model T) you're producing. Even if that is a good idea here, the standard -- pay workers enough to buy the output -- is a bad one, because the output could be anything from a cheap consumer good to a megaproject. [1] which, full disclosure, I've made myself: http://blog.tyrannyofthemouse.com/2012/01/mr-ford-meet-boeing.html http://blog.tyrannyofthemouse.com/2012/01/mr-ford-meet-boein... [2] even under general mechanisms by which "price tends to cost" and intelligent uses of the labor theory of value, a worker might be contributing a valuable input (engineering design), or a not-so-valuable input (cleaning services for the offices).
- 34679 7y agoBoeing never intended to build a product that an average American can afford. Ford set out to do exactly that.
- gshdg 7y agoBut it is in Boeing's interest to pay workers enough to increase demand for jetliners -- that is, enough that they can afford airline tickets with some frequency.
- philwelch 7y agoNot necessarily! If Boeing decides to pay an employee an extra $50/month, and that employee saves all of that money and spends it on airline tickets, after the money that goes to the airline, the airline workers, the oil companies, and the airports, a small fraction of it returns to Boeing. Whereas if they didn’t pay them that extra $50/month, they would retain 100% of it.
- iso1631 7y agoIf you pay an extra $50 a month, it forces all the other companies to pay an extra $50 a month, and you get a fraction of everyone's extra $50. If you get 5% of the average $50 a month extra, but only employ 1% of the workers, you make money
- philwelch 7y agoIf you only employ 1% of the workers and you pay an extra $50/mo, the rest of the market isn’t necessarily going to keep up with you. You’re probably just gonna end up hiring the top 1%. (Which also had a lot to do with Ford’s success, to be fair!)
- Talanes 7y agoAnd if every business does this, they all retain the $50/month and none of their employees afford anyone's products.
- maest 7y agoAnalogous reasoning suggests that, if you have a 401k and thus probably own some shares in, say, Starbucks, then it's in your interest to spend as much as possible at their shops.
- 7y ago
- Wowfunhappy 7y agoIt’s disingenuous for another reason—outside of the extremely rare private jet, airplanes aren’t intended for consumers. Seats on an airplane are. So the good-faith version of the argument would be: people who work at Boeing should be payed enough to afford going on vacation by airplane. (I’m not saying I agree with that statement, as it’s an odd benchmark for other reasons. But it isn’t ludicrous.)