Vince Cable's Graduate Tax

The latest news about university funding is Business Secretary Vince Cable's proposed Graduate Tax, where graduates pay a tax dependent on income after graduation and the money is channelled directly to universities.

Left Foot Forward covers the story here, however the focus is on Cable himself and on whether or not he will keep his campaign promise of scrapping tuition fees, rather than a discussion of whether this proposal is the best way to fund higher education or of the wider implications of the policy.

LFFs analysis is based on the assumption that:

The fair solution is to abolish tuition fees and ensure that graduate contributions are based on actual earnings in the real world, rather than sticker prices in prospectuses, which are based on guesswork.

However, little to no thought appears to have been given to whether or not this is the best way to fund higher education and what impact this will have on decisions made by students and graduates alike.

I think this is the appropriate time for a full disclaimer:

I am a university graduate; I have an engineering degree and a pretty good job; I had the privilege of attending university without the need to pay any tuition fees whatsoever (I do have a substantial loan to pay back, but that was only to cover living expenses).

In principle I have nothing against the idea of getting the people who have benefited most from a service to pay the most for it.  My concern is not simply out of self-interest of not wanting to pay any more tax.  My concern is in how the tax will affect people's behaviour.

For example how many potential students of engineering/medicine/law/etc. will choose to study history or teaching instead, figuring that they're likely to earn less in those professions therefore avoid the graduate tax?  How many potential students will decide not to go to university at all?  (This may or may not be a legitimate concern, considering that UK universities are already grossly over-applied for).  What will be the impact of these decisions on the wider economy in the long term?

Maybe the numbers are small and these concerns of little consequence.  What troubles me is that I haven't seen any analysis of the proposals that even considers these questions.