The signaling isn't just for intelligence and grit (although those are factors); as Caplan argues in The Case Against Education, a college degree also signals capacity for obedience and conformity.

https://press.princeton.edu/titles/11225.html

I don’t buy this. Are saying if you get college a degree in Chemical Engineering you wouldn’t be far more productive as a Chemical Engineer than if you only went to high school? Or in say, physics or chemistry or botany?

No the commenter is saying that having completed a degree in chemical engineering serves as a signal to your productivity. It does not in fact determine how productive you actually are.

Observation: college graduates are more productive laborers in the work force.

Hypothesis A: college coursework actually increases a student's future work productivity.

Hypothesis B: the least productive laborers are unable to be admitted to a college, or unable to graduate. The admissions and graduation filters thus raise the productivity average in the pool of graduates.

Hypothesis C: businesses with the most productive workers tend to prefer college graduates. Workers without degrees appear to be less productive because the only jobs available to them have inherently lower productivity.

Hypothesis D: productivity is measured differently for some workers, based in part on whether they have college degrees.

Hypothesis E: some workers are able to influence the measurement of productivity, such that productivity originating from other workers is reassigned to them, and those cheating workers tend to have degrees.

Hypothesis F: workers with degrees tend to have debts, and are forced by necessity to be more productive in order to first pay them off, and then later save enough for retirement with a shorter savings window.

Hypothesis G: workers with degrees are more productive because the past correlation generates an expectation that they be more productive.

The observation could have complex cause, and all of the above may be true.

Almost everything interesting that is above chemistry[1] in the science hierarchy has complex causes. It would be a good thing if people in those fields would get over their physics envy and try to communicate this fact to the public better. The number of "this flawed study suggests that some small minority of people might be a bit more healthy if they restricted their sodium chloride intake" becomes in the general population "eating less salt is good for you". And foods can somehow be labeled "low salt" when the manufacturers lower the sodium and jack up the potassium (both are salts) level to keep the taste close to the same.

[1]https://xkcd.com/435/

Well, I understand that the GP is saying that expecting the same person to have the same productivity as a chemical engineer after and before a graduation on that area is ridiculous.

Claiming that high education has only signaling value is ridiculous. (It's even more ridiculous than claiming it has no signaling value.)

There are varying extents to which signalling is a good explanation and employers have more fine grained information then just degree or not. As usual the literature is less univocal.