I confess to feeling deeply ambivalent, not to say conflicted, by the student fee debate. If the country will not stomach tax increases to pay for higher education (wishful thinking aside) and the deficit cannot be cut in different ways (a bit more debatable, but where would you cut?), then higher education has to be reduced in quality, reach or both unless other funds can be found. Graduate taxes or deferred fees are the only remaining options.

Whilst the evidence suggests top-up fees have not deterred working class students from entering higher education, the increases the Government is allowing will be of a significantly different order. Changes to the terms of repayment may soften the blow, particularly for the poorest students. I suspect most would-be students will take the (delayed) pain, hold their noses and head off to university when the time comes. There will, though, I believe be some impact: both on the types of course that students study and the numbers coming from particular backgrounds. Some students WILL be put off by the debt and those will be coming from poorer backgrounds.

However, I doubt whether the effect will be as strong as is suggested by the level of public concern, and abolition of the education maintenance allowance may be more serious: reducing the numbers of students proceeding in education beyond 16. I want, however, to pick up on something Peter Wilby has argued in an excellent piece which is worth quoting extensively. He claims fees have increased access because they have funded an increase in access to higher education. It emphasises the progressive case for tuition fees:

"Far from excluding those from poor homes, the introduction of student fees has been associated with a sharp increase in their participation. This is because universities have been able to admit more students without the cost to the exchequer being prohibitive. In the late 1990s, before fees, universities – like grammar schools of old – were vehicles for passing middle class privilege down the generations. The 11-plus was transmuted into an 18-plus. The post-1960s expansion of higher education benefited the less intellectually gifted middle class children and, more laudably, middle class girls. During those years the chances of a middle class child getting to university rose faster than those of a working class child. The gap between the social classes increased.

"In 2009, however, young people from disadvantaged neighbourhoods were 50% more likely to get university places than they had been 15 years earlier, while those from advantaged areas were only 15% more likely. At last university expansion was benefiting the poor, giving them a smidgen of hope that they could climb the social ladder.

"I repeat: a smidgen of hope. The class gap in participation rates remains wide. In poor areas, fewer than one in five go to university, against more than half in affluent areas. Equally important, entry to elite universities – those that all but guarantee entry to sought-after careers – still carries the heaviest social bias."

There is a lot of sense being spoken here but it is particularly important to note some of the detail in the reasoning. Firstly, reasoning that tuition fees are progressive is dependent to some extent on the expansion of places being maintained or extended. There are two potential problems: 1. the proposals may depress demand for places reversing the expansion and 2. because the government bears the upfront cost of fees they retain an interest in not extending the number of places further or, if they need to save further money, reducing the number of places.

The second problem is the one that particularly concerns me. Entry to higher education is not a binary concept. You are not either in or out. If you gain entry to an elite institution you are 'in', but there are degrees of 'in' even amongst that elite. How in you are depends on whether you get into Oxbridge, the main London colleges, the Russell Group or A N other university. There are degrees of eliteness within the elite and your likelihood of getting in and up within that elite are heavily dependent on social background. Higher education is very stratified: the 'good' employers recruit from the 'good universities' who recruit from the 'good schools'. This cycle boosts the reputations of all of the winners: in league tables and beyond.

One might hope that this was a meritocratic process, and to a degree it will be, but there are serious flaws in the way that those ideas of merit are measured and constructed (see, for example, here on A-levels and here on the idea that 'commitment' is related to suitability to practice). The question is not whether individual institutions discriminate (some may but proving it will be difficult). It is much harder to see an argument against the claim that the system discriminates.

In a nutshell: we allocate elite places to the best public school children but only the very best state school children. The claim that this is because the benefits of public school education are reflected in better performance at university appear to be wrong. If Sutton Trust research shows that a student with BBB from a state school will perform as well as a student with AAB from public school then we need to change our admission criteria to make sure we recruit those who will do best at university not simply those who have done best at school. For genuine social mobility to take place and, from the profession's perspective, for the members of the Bar and Solicitors professions to be recruiting the best from the best, there needs to be change (see Alex Massie's very interesting blog for one approach which bears more scrutiny than I suspect it will get).

Unless there is a genuine rethink on entry criteria, which challenges the biases built into the current system, then the fee reforms are likely to make this worse. One particular problem is that students from poorer backgrounds are likely to choose institutions closer to home. There are three reasons: their local university may be easier to get into (depending on where they live of course); they may believe that a degree is a degree (and so not care enough about where they go); and (critically) studying locally will be significantly cheaper. Fees are only one barrier to education. Living costs are a more pressing and similarly sized barrier: studying locally ameliorates that problem. It will, however, tend to exacerbate the stratification of higher education.

If this happens, employers need to be encouraged more firmly, and on the basis of enlightened self-interest, to recruit more broadly than they currently do if this stratification is not to be more damaging than it needs to be. There may be things universities can do to encourage this, but both they and employers will be difficult to persuade. To advance high quality, merit-based education, there are plenty of battles to come.

Professor Richard Moorhead is deputy head of the law school at Cardiff University. Click here to visit his blog, Lawyer Watch, and click here to follow Richard on Twitter.