Students are getting a raw deal at university because money is being drained from teaching and research to support the government's widening participation (WP) drive.
Research shows that the government's failure to fully fund its policy of increasing the numbers of students from poor backgrounds forces universities to subsidise access from other budgets.
The study, Determining the Costs of Widening Participation , by Universities UK and the Higher Education Funding Council for England, shows that universities are being shortchanged by nearly £120 million a year.
It suggests a cash premium for widening participation of 35 per cent on top of the average amount per student. The premium next year will be 10 per cent. On this basis, next year's premium of £47 million should be £165 million.
The study concludes: "This implies that (higher education institutions) are effectively cross-subsidising their commitment to supporting the government's WP goalsI from other income and activities.
"In discussion, the two pilot universities suggested that the WP subsidy was mostly coming from academic and other staff time, either unpaid or diverted from other activities (including research and scholarship).
"More worryingly, there was some suggestion... that growing additional demands for student support from WP entrants were impacting on the overall quality of provision for all students."
The findings fly in the face of government claims that it will fully fund the cost of the student expansion required to meet its target of half of all 18 to 30-year-olds benefiting from higher education.
Ministers have made it clear that expansion should bring in more people from disadvantaged backgrounds but, as the report indicates, they appear unwilling to pay for this strategy.
UUK chief executive Baroness Warwick said: "There is now recognition in all quarters that to recruit, retain and transfer non-traditional students effectively into work or further study means additional costs."
The study, carried out by PA Consulting, is based on an analysis of the costs of recruiting and retaining students from poor backgrounds at two unnamed English universities. One is a post-1992 institution with nearly 14,000 full-time equivalent students and the other a post-second world war institution with 6,500 full-time equivalent students.
It costs £1,884 more for widening participation students at the new university and £1,475 at the older institution. The overall cost of widening participation at the new university is nearly £9.7 million and just over £1.9 million at the older institution.
The difference in cost reflects the larger numbers of widening participation students at the newer university, which devoted proportionately more staff and money to supporting students in their first term and throughout their course.
The older institution focused much of its effort on attracting disadvantaged students to the institution through raising aspiration and other pre-entry activities.
The study calls the newer university's strategy more "mature and efficient" as it was already attracting significant numbers of widening participation students and ensuring that they did not drop out.
PA Consulting interviewed about 20 staff at each university. Their widening participation duties included training staff in access processes, developing links with schools, offering classes on study techniques, assisting students through clearing, and providing personal and academic support to students in their first term.
It concluded that, while it would be misleading to draw too many conclusions from the sample, the methods employed for widening participation were essentially robust and "portable" across a wider sample.
* Northern Ireland's apparently impressive lead in widening access in the UK slumps in terms of higher education students from the most deprived areas, an Ulster University expert has revealed, writes Olga Wojtas.
Bob Osborne, professor of applied policy studies, said a higher proportion of the province's population were skilled working class, semi-skilled or unskilled, which meant its higher proportion of entrants from these classes was only to be expected.
But its proportion of entrants from semi-skilled and unskilled backgrounds was less than 10 per cent in 2001, compared with more than 10 per cent in England and Scotland, and almost 12 per cent in Wales.
Professor Osborne was speaking at a widening access seminar hosted by the Northern Ireland Assembly's committee for employment and learning. This was the first of a series of regional seminars on the findings from the Universities UK report, Social Class and Participation .