Are income-contingent loans for higher education feasible?

There is considerable discussion, in particular in the UK, about making education loans contingent on realized income after completion of education (and those who do not graduate have the loan paid by general taxes). On the face of it, it makes much sense as those who reap the most benefits from their education pay the most for it. But the bigger advantage is thought to come that such a scheme would help overcome risk-aversion, educational outcomes are risky after all, and thus help more people to get higher education. Also, because education then does not require upfront payments, it helps alleviate liquidity problems young people typically face.

Elena del Rey and María Racionero see that this is not that easy to implement. Indeed, some people would lose if such loans would be implemented, foremost those who do not need loans (but this can be reversed if they are very risk averse) and those who are little risk averse. Regarding education subsidized from general taxation, those who can afford private education are again opposed, as well as those who are less able to acquire higher education. If one were to choose between the two financing schemes, it would all depend on the political power. If we assume everyone has a vote of equal weight, then all boils down the level of risk aversion of the population, the distribution of which we know very little about, and the aptitude to education, which is rather uncertain at the individual level. So in the end, we are not much wiser.