When will those who commission scientific research and support undergraduate teaching in such subjects do their sums and pay the real overhead costs ("Clear improvement?", THES , September 26)?

We would scarcely need a research assessment exercise if they did. Everyone knows the overhead for research projects is between 80 and 120 per cent. The consequences of not paying the bill are clear. Most university premises and facilities are tatty. When did you last see a lecture theatre clock that told the right time? How much equipment is unused because the research grant came without a maintenance contract, or with no provision for breakdowns, or finished two years ago - well short of the correct depreciation for an expensive capital item.

By giving away equipment on day one, research granting bodies in effect write off millions of pounds instead of correctly depreciating the asset they paid for. The Treasury or National Audit Office should step in to stop this.

Sir Gareth Roberts' proposal to calculate the full economic cost of a project does not bite the bullet. He suggests paying 60 per cent. Where is the other 40 per cent supposed to come from? Teaching?

ÖйúAƬ

Research councils distribute about ?1 billion among universities. The Medical Research Council and others distribute nearly ?1 billion.

Government departments spend well over ?1 billion on commissioning research. The total missing money must exceed ?1.25 billion, and that does not include underfunding from sources such as industry or the European Union.

ÖйúAƬ

How much gets awarded after all the agonies of the RAE? Only ?850 million. The overall deficit is about ?500 million. This equates to a shortfall of more than ?5,000 for each and every research-active university scientist.

Scientists who emigrate take their expensive training with them. They may cite higher salaries and lower living costs as reasons, but they also mention the quality of facilities and infrastructure that enable them to do their job properly. The UK might trumpet its ability to squeeze value for money in terms of research published and citations gained, but it is clear that we are slipping. Postdocs who stay in the US have a higher publication and citation rate than their brethren who return. The current funding system is topsy-turvy. Does anyone have the will to sort it out?

Alan D. B. Malcolm
London, SW5