He's not really all that wrong, though. Fundamentally, the problems are that there are more potential employees than there are jobs, but that there are fewer "really good" employees than jobs. Hiring processes are meant to identify the few among the many, but they use shortcuts.
40 years ago, when only 15% of 20-somethings had a college degree, that degree was a pretty good indicator of "ambitious, works hard" which made the degree a good hiring litmus test and enshrined it as a ticket-to-a-job. Today, if a hiring manager has to choose between otherwise identical candidates, they're likely to take the one with a degree over the one without.
Today, when 30% of 20-somethings has a college degree, it has lost a lot of its value as an indicator of "ambitious, works hard," and hiring managers have had to move on to other indicators. Internships. Co-ops. Extracurriculars.
The problem is that as soon as those indicators become known, people start trying to game the metrics. It's like a cargo-cult version of professional development. I can't tell you how many kids I've heard sign up for this-or-that school club just because they believe they have to have some extracurriculars. Not because they have any actual interest in [whatever], or any intention to actually attend meetings and events, but just because they have to have that line on their resume. Conventional wisdom is that these things will help you get a job; student hears these things will get you a job.