Hiring directly based on IQ is illegal.

This is not correct, in the U.S. states where I've practiced employment law.  Are you talking about a particular state?

If this is your interpretation of the US Supreme Court's decision in Griggs v. Duke Power Company, I would disagree.  

Nothing in the Act precludes the use of testing or measuring procedures; obviously they are useful. What Congress has forbidden is giving these devices and mechanisms controlling force unless they are demonstrably a reasonable measure of job performance.

401 U.S. 424, 436 (1971).  What had happened was, Duke Power enjoyed its explicitly racially segregated work force and wanted to keep it, but the newish Title VII of the Civil Rights Act of 1964 said that was illegal, so Duke imposed an intelligence test on certain jobs it wanted to keep white but which didn't really require high intelligence-as-tested.  This had the effect of disproportionately excluding Black applicants, who apparently didn't do as well on the test for reasons that are not germane to the holding.  

Held: If a facially neutral job requirement has the effect of discriminating based on a protected category, the employer has to show that the requirement is "a reasonable measure of job performance"--with a strong undertone of "How stupid did you think we were?" from Chief Justice Burger, writing for a unanimous court.

mindless labor like farming

Consider being more specific, since produce-picking is arguably mindless labor but "farming" isn't necessarily. Farmers do things like deciding what to plant, knowing timings / how weather affects them, diagnosing and fixing problems, doing marketing and sales, and hiring and managing other people if the farm is big enough. In your hierarchy, I would put most farmers (as-in owner-operators of farms) somewhere between skilled labor and petit bourgeoisie.

Attempting to break into the middle class can be risky due to the sticker price of college plus the lost wages.

My experience has been that the myth of expensive college is a bigger problem than the actual cost of college. The difference between my yearly pre-college income and post-college income was enough to pay off my entire student loan the first year. Of course that depends on treating college like an investment (get the training and credentials you need at the lowest price) instead of following the bad "self-fulfillment" advice, but I think that's a different problem.

The problem with intellectual labor is it's hard to tell whether someone is doing it right. If there were clear criteria for success then the job would have been automated away by now.

This isn't necessarily true. You mention sales later, and I think that's a great example. I can trivially write a program that looks at your sales data and determines if the trend is positive, but I can't write a program that calls customers and convinces them to buy my product. The actual process and verification are frequently completely different processes.

Even with something much harder to verify like management, this still holds. It's true that it's hard for me to tell if someone is a good manager, but it's still much easier to check results (what does their team's turnover look like?) than to know exactly what steps they should follow to make that metric better.

Moreover, if you knew what the right thing to do was then you wouldn't need to hire someone else to do it.

I don't think this is true. It's common to know exactly what you want done but not have time to do it. For example, Sergei Brin knows how to program, but he still needed to hire thousands of additional software engineers.

When white collar workers collaborate in teams it is impossible to how how much each individual employee is worth.

It depends on how big the gap is, but I'd say this is difficult but not impossible.

Credentialing. It's hard to measure if someone is a good electrical engineer but it's easy to measure if someone has a degree in electrical engineering or used to work for Facebook.

For what it's worth, this is sort-of changing in the software field. It turns out credentials aren't good enough and it's more effective to directly measure how good someone is at writing software (either with tests or by looking at code samples). Unfortunately, most places seem to do this as a two-step gate (you need credentials and to pass the "can you do the job" test), but I think that's more of a path-dependence thing than direct usefulness (path dependent because historically the most competant people got software engineering degrees, so if you don't get one now it implies that you can't, even if in a world without that history, not having a software engineering degree wouldn't mean anything).

Great post! Love the point that "it is common middle class myth that you can work your way up the corporate ladder to the top". Seems it is a mindset problem with middle class there. It would be interesting to elaborate on the ways to change the mindset and jump out of the middle class trap.    

What are the middle class? the 50%? 60%? In the first place, middle class is not well defined. Perhaps Orwellian middle class is neither middle nor classy. It just define middle class as the 18% of outer party, between 80% of Proletarian and the 2% of outer party. 

In term of wealth, guess which group is growing faster? The top 20% earners, defined by having more than $200,000 income. The middle class, in this case earning anything between $40,000 and $200,000, is shrinking fastest. If so, the definition will change "middle" class as the fastest growing demographic group.

The whole article perhaps has a fatalistic outlook, it is a we versus them analogy. Had not those entrepreneurs, working 60-70 hours per week, longer than most average person, is considered a working class as well?

And Microsoft did own by the public. The fact that Bill Gates doesn't own Microsoft, is a testimony of a Public Market. It is not called "Public Limited Company" for no reason. It is now majority owned by the public, whose face unknown and many.

Yeah, I believe the army checks IQ, and the whole IQ problem, where people with low intelligence can't perform certain jobs, it's sad.  That's probably why I'm stuck running my company Drywall Repair Columbia.  It's just too bad that IQ can't really be improved per se, maybe we'll find a way to fix that in the future, who knows.