Let's consider the distributional effects of giving college aid to 10% of the population vs. 50%. If the 10% were people who would not have otherwise gone to college, it looks like a good move to create income mobility and increase economic growth which will ultimately create more even tax revenue. If the 10% would have gone to college anyway and taken loans (and easily paid them back with their top 10% earnings), the aid starts to looks regressive instead. But when we are talking about 50% of the population, things seem different.