Margins matter. Any economist could have told you so years ago, but now the entire academic community is learning that reality as a consequence of the research assessment process.

The RAE, which began in the mid 1980s and is due for its next round in early 1996, is simply a mechanism for distributing research funding to universities. In England this year RAE funding is worth Pounds 600 million - less than one fifth of the total distributed to institutions by the ÖйúAƬ Funding Council for England and less than 10 per cent of universities' total income. Yet that comparatively small funding element has had a profound effect, helping to transform the culture of an academic profession that jealously guards its autonomy.

That impact is a consequence of the funding environment of the last decade. In a period when money is tight, with a constantly declining unit of resource, good research assessment outcomes have offered - or appeared to offer - most institutions' best chance of reversing or slowing that decline. Marginal money maybe - but margins are where jobs are lost and tenured appointments give way to short-contract part-timers.

Distrust has been entirely natural. The past decade and a half has seen a proliferation of performance indicators, many of questionable validity. The mania for quantification, which reduces complex processes and institutions to a single index, and runs in parallel with and reinforcing a government mindset in which "efficient" means cheap, has done immense damage across swathes of national life.

ÖйúAƬ

The community is also well aware that indices rapidly lose their original meaning once used for funding purposes. Academics are critical analysts by training, so it is hardly surprising they have used their skills to deconstruct the system. That deconstruction leads logically to system playing by institutions and cynicism from individuals. The harmful consequences of games playing - inequalities between research "stars" and others, a destabilising transfer market, the rush to publish and loss of collegiality - are clear from the articles we have run over the last four weeks.

But this is only part of the story. Just because two things happen at the same time, they are not automatically connected. The RAE's impact on the academic psyche has been so significant that it is often blamed exclusively for phenomena such as the explosion in academic publishing, which can also be observed in RAE-free zones such as the United States and Canada.

ÖйúAƬ

Nor are the funding councils entirely to blame for RAE-related problems. Selective funding - the sole reason for RAE - is not their invention, but the Government's. Once it was ordained, it was their task to find the best and fairest means to accomplish it.

And before damning RAE we should consider the alternatives. It is too easy to forget that funding was always spectacularly unequal. Do we really want to set the allocations of 1984 in aspic? The RAE is costly in terms of academic energy and time. But would it really be better to leave it to the Man in Whitehall? Much time and energy is taken up at departmental as well as institutional level as each subject-group bids for funds. But would it be better to have a system determined by institution only, leading to the much-touted "superleague" of leading research universities attaining not just their current dominance, but a total monopoly of research?

The answer to all three questions is surely no.

Nor can the funding councils be accused of failing to respond to the views of the community as the process has evolved from its frankly unsatisfactory 1986 incarnation. Peer review, vastly preferred to other forms of assessment, remains the rule. The crude publications count, leading to salami publishing and quantity against quality, has been dropped. And while the 180 or so pages of criteria issued last week may have induced groans of horror from administrators already over-burdened with reading matter, their sheer bulk testifies to a determination to make the system as open and comprehensible as possible.

Which isn't to say that There Is No Alternative. There always is. But rather like Churchill's description of democracy, RAE is undoubtedly the worst system of research allocation ever devised apart from all the other ones.

ÖйúAƬ