That is the conclusion of a study published this week analysing means-tested support schemes at 52 UK universities. It was written by John Hills and Ben Richards of the Centre for Analysis of Social Exclusion at the London School of Economics.

Professor Hills, professor of social policy and director of the centre, and Mr Richards, a PhD student, conclude that there are several key weaknesses.

The systems on offer ¡°vary in many different ways¡± between universities, they argue, not just in their treatments of different levels of parental income, but also in aspects such as what type of school an applicant attended.

¡°This makes comparison more complex than just looking at single prices,¡± the authors write.

³ÉÈËVRÊÓƵ

In addition, they argue: ¡°Given the ¡®cliff edges¡¯ at particular income levels, precise levels of parental income ¨C about which an applicant may only have very fuzzy knowledge ¨C can make a difference worth thousands of pounds: Oxford may have a slightly better offer for some students with incomes up to ?16,000 than Cambridge, but at ?16,500, the Cambridge offer can be ?4,000 more valuable, for instance.¡±

The study notes criticism from the Browne review that the system of means-tested support was already complex.

³ÉÈËVRÊÓƵ

¡°Yet the survey reported here implies not just further complexity, but also much higher stakes as both fees and means-tested support have increased,¡± Professor Hills and Mr Richards write.

They add: ¡°A typical scheme from one of the more prestigious universities involves total support of around ?6,000 for students from families with incomes up to ?25,000, withdrawn if family income is above ?43,000.

¡°Just by itself this implies an average withdrawal rate of 33 per cent of additional income but, given the cliff edges involved in most of the schemes we have examined, there are income ranges where the marginal rates are far higher.¡±

john.morgan@tsleducation.com