If those older workers and others are equally or more skilled than the younger ones implied by your question, smart companies will realize the discrepancy and hire them.
In addition, if older workers know that firms place a "high value on undergraduate-level CS theory," they should probably spend some time learning. . . undergraduate-level CS theory.
You can in fact see this in action in other areas—for example: http://www.economist.com/node/17311877 .
The invisible hand of self-interest only works if smarter contenders actually appear in the marketplace. If there is some invisible hand of stupidity (and groupthink) that affects all companies above a certain size, then we are hosed.
Why is it that all big organizations are almost universally dilbertesque? There must be some anti-nootropic effect that occurs above a certain threshold of organizational complexity.
http://www.ribbonfarm.com/2009/10/07/the-gervais-principle-o...
This coincidentally depresses the marketability (for a certain class of job) of older workers and demographics less likely to achieve a degree from a top N computer science program.
Certainly you could read between the lines and say that observably favoring e.g. Stanford 2012 CS majors effectively disadvantages some protected minorities (age, gender, race). By placing the emphasis on a particular credential rather than on demographics a hiring company gets a level of plausible deniability from such claims.
Disclaimer: I have an MSci in CS and I'm only 30, so I'm not too worried about the things implied above. Yet.