The book is a fast and mainly interesting read. Unfortunately, I found the book a bit too much focused on a few characters, rather than really trying to narrate the process itself. It was almost gossipy rather than journalistic (and not even close to academic.) Clemens seems to want to talk about the death of manufacturing jobs--- he explicitly talks about how disassembly and scrap is a bigger market than building new factories. But, in my mind he hangs too little content on his narrative, and the result is unsatisfying.
I was caused to speculate a bit about whether the search for lower labor costs is partially driven by a plateau in automation capability. In the grand scheme of things, laying off a few hundred workers in the 2000's by moving production to Mexico is trivial compared to the order of magnitude larger drop in workers, at the same plant, in the previous few decades.
Reading the book also reminded me a bit of news stories about the possible cap on debit-card fees. Banks were quoted as saying things to the effect of "if we lose that $N million in revenue, we will have to cut K jobs to compensate." From the Star Trib:
In court documents filed last week, TCF controller and assistant treasurer David Stautz warned that a 12-cent limit on debit-card fees would reduce the bank's annual profits from $120 million to $70.4 million, a 40 percent drop. To fully mitigate the impact, the bank would have to eliminate 1,777 jobs -- or 28 percent of its total workforce.
Anybody using this reasoning should not be trusted with your money. Unless TCF is getting out of the debit card business, TCF will have the same responsibilities after the regulation as before. Either those jobs could be cut now--- saving the stockholders of TCF a cool $50m--- or the work will still have to be done later. Stautz's argument is either a blackmail threat or sheer nonsense.
Now, I realize that it is not quite that simple. It's hard to figure out, in a large business, how many people you actually need working on what. So, the amount of money you have to pay people can be an indicator of how many people you should be hiring. Moreover, the level of staffing is one of the few things that can be changed (reasonably) quickly--- it's easier to fire people than to reduce real-estate expenses, close branches, power off data centers, etc.
Still, the common thread I see here is that employees get treated purely like costs. Sun did this too whenever it sent out an email begging people to take their vacations (and imposing a mandatory vacation week) and I did not find it motivational. But employees are also what produce income, and companies like Cisco are right to emphasize that by reporting revenue-per-employee. If you don't know how employee efforts translate into income, it doesn't matter whether you're paying Facebook wages, Detroit wages or Aguascalientes wages.