I want to second this point. The percentage of young adults (this is the demographic that represents most fo the current community college population) living below the poverty level is about 40% [1]. So even in the absolute worst case scenario where the Pell grant only applied to those below the poverty level and all those above that level started taking advantage of the free community college your efficiency could drop by a factor of two. In reality, there is a significant fraction of those below the poverty level who don't know about such Pell grant opportunities and the complexities of federal student aid and would never apply, but would jump at the opportunity for a free education. Also, a more realistic assessment of the efficiency question would require a better picture of how many people below the target of the Pell threshold would start taking advantage of the program. I would suspect that only a small fraction of people from the top 10% income bracket would pursue this, since they have other opportunities available. So lets say that we actually have something like a 50% reduction in efficiency to get the money to the target audience, and a side effect of a few tens of millions of people outside the target audience getting their education subsidized. There is obviously room for improvement, but I would support that scenario.

[1]http://www.ihep.org/sites/default/files/uploads/docs/pubs/br...