In general, you're right. But I think you'd be hard-pressed to defend the claim that there is no causative relationship whatsoever flowing from college education to employment. I don't think it's terribly controversial to claim that employers pretty universally view more education as a positive factor on job applications, all other things held equal.
edit: to elaborate further, I think you might be bringing up the wrong fallacy here - at least, if you're looking to highlight the problems with representing program participants by the general population of high school-educated people.
What I'm saying is that it's not obvious that if the student who was admitted to Harvard decided not to go, he'd make less than the student who was admitted and did go.
edit: to elaborate further, I think you might be bringing up the wrong fallacy here - at least, if you're looking to highlight the problems with representing program participants by the general population of high school-educated people.