Not seeing the downside yet. You want to cultivate a pool of bright, dedicated people to work for you one day. You give them a tool - free of charge - for them to play with, develop their skills, maybe use the tool in ways no-one anticipated, let their creativity run free, maybe one of them will produce a product you'd be prepared to buy or license from them, and then offer them a job. Can you point me to a loser in this deal? It's not like a free software advocate, i.e. a Blender user, couldn't produce an impressive CV to show the hirers at Pixar, right? When you have to choose between 2 applicants of equal merit EXCEPT one them knows how to use your tools, and the other doesn't, who do you choose? Who do you choose when the Blender user is *slightly* better than the Renderman user? Of course, someone *really* dedicated will have skills in both packages.
Apple do it. Microsoft do it - although their motivation is less to get you to work for them, than it is to advocate the purchase of their software, wherever you work. There is (or should be) no legal reason that schools can't install free alternatives (and some do just that). They make their decisions based on a lot of factors - the perceived market for their students' skills, the bias of selection committees, ease of use, and outright bribery in some cases - but free software needs to compete on more than its merits, unfortunately.
Show me an easy installation package (LibreOffice ticks that box), a series of relevant templates that meets the teachers' needs (not sure, haven't seen any, yet), and interoperability, and I'll advocate free software. Sadly, it misses out badly on the third criteria. Fortunately, MSOffice since 2007 has been less usable than before, and the free alternatives have become more attractive. I've had customers select LibreOffice over MSOffice 2010/2013 when upgrading, because they just want the old interface (and they've "lost" the Office 2003 installation disc).
All that said, I'm going to try Renderman.