Ryan Avent has some fun with Graeber.  Click for the article but here is a summary.

Graeber "The ruling class has figured out that a happy and productive population with free time on their hands is a mortal danger".
[skip]

Avent (my bold)

"Employers had to retain such workers—had to pay them a wage sufficient to keep them on the job despite its dreadful tedium—because the machines of the era lacked the manual dexterity to complete the required tasks, and so a line of human machines was the only way to make the highly productive assembly-line system work. As technology evolved, however, automating routine tasks became ever easier. And the high wages needed to compensate labourers for the soul-crushing repetitiveness of their work gave employers every incentive to automate routine tasks as soon as it was technically feasible."

1) Graeber's "ruling class" is a mystical entity,the illigitimate child of the Zeitgeist and the species being.  But that doesn't mean there aren't a lot of bullshit jobs.

2) assembly line work is very boring. The wages were very high.  It is natural to guess, as you do, that the high wages were a compensating differential.  However, you don't have to guess. There is a huge literature which attempts to test your hypothesis.  Your guess is incorrect -- it is massively rejected by the data.  It is easy to see that high wages received by people without fancy degrees in manufacturing (especially heavy manufacturing)are definitely not just compensating differentials.  Such jobs are eagerly sought.  Quit rates are very low (note high wages because quit rates which are fine for other firms are too high for assembly line operations are *not* compensating differentials but rather efficiency wages).  I think it is safe to say that almost everyone who has studied the issue dismisses your compensating differential argument. The Chicago view is that the differentials are unobserved human capital.  The case against your hypotheis is so strong that the consensus that you are wrong is probably about as strong as the consensus that human activities contribute to global warming.  In any case it is a super highly studied issue and some familiarity with the massive Econmic literature might add to the value of The Economist.

Avent again

" How does the ruling class co-ordinate all this hiring, and if much of the economy's employment is useless in the first place why not just keep them on during recessions?"

3) You mock the idea that managers ever do anything which doesn't maximize profits and shareholder value.  This is a declaration of faith.  No one who studies actual firms believes this.  The profit maximizing firm is purely theoretical concept like an ideal gas of a frictionless system.  Another possibility noted by Northcoat Parkinson is that administrations tend to expand to fill the allowed space.  Again this is a testable hypothesis. If the managers are necessary, then firms making similar products will have similar numbers and there won't be a pattern of higher profitability of firms with a lower proportion of paper shuffles (really power point presenters) to production workers. That is you predict that GM and Toyota will be similar in number of layers of management and economic success.  Or how about Ford ?  When Iacocca left to save Chrystler a hated (and forgotten) Phillip Caldwell came and promtly laid of 30,000 headquarters staff.  What valuable management services did Ford then lack ? Which US car company didn't go bankrupt.
According to you the whole movement towards downsizing and delayering in the 90s was silly.  The layers and size are always presumably efficient.  The productivity growth, increase in accoutning profits and increase in market value must be an illusion (or maybe it was the Zeigeist doing it ?).

4. The pro market view is schizophrenic.  The presumptions is that people are the best judges of their own interest, that economic agents are rational (maybe not the full rational expectations) and that ordinary people are totally confused.  The workers who say they really really want manufacturing jobs don't know what's good for them. The workers who say their job is a waste of their employer's money don't either*.  The vast majority of the supposedly rational agents think you are full of it.  You argue that they understand the economy (needed for Pareto efficiency) and that they don't understand it at all.  Your argument is based simultaneously on self abasing esteem and utter contempt for the same set of people.  It is both conventional among economists and inconsistent.


* I am one of those workers.  But I don't suffer from the soul destroying tedium of doing pointless things.  I blog.