The most depressing sums in science: adding up numbers in the brain drain

Martin Rees: 'The loss of even a few "stars" impacts on the quality of our national scientific enterprise'

Science is not a monolithic profession, nor a single constituency; certainly not an effective lobby. As with other long-term issues (energy and environment, for instance) it's best if national science policy isn't the subject of strident debate. But the "downside" is that it then slides down the political agenda, except when attention focuses on some "scare" or crisis.

Science is not a monolithic profession, nor a single constituency; certainly not an effective lobby. As with other long-term issues (energy and environment, for instance) it's best if national science policy isn't the subject of strident debate. But the "downside" is that it then slides down the political agenda, except when attention focuses on some "scare" or crisis.

The Government's White Paper only rated a few mentions in yesterday's newspapers. But it deserved more. Its most eye-catching proposal was a new initiative to attract and retain top talent. We are an international market, and are losing out, even though we have strong advantages which should make science one of the country's distinctive strengths.

There is a net flow of top-level academic talent from this country to the US. The proportion of Fellows of the Royal Society working in the US has increased over the last 20 years; the top US scientists who have come to this country would make a shorter list.

The "defection" of leading creative artists, performers or designers would be seriously detrimental to UK life. Likewise, in each main field of science, loss of even a few "stars" impacts on the quality and cost-effectiveness of our national scientific enterprise. It also sends a negative signal to the next generation. If the outstanding discoveries highlighted in the media are made in the US - worse still, by UK scientists who have "defected" to the US - what message will it give to young people who are contemplating careers in science and technology?

It's in the UK's interests to match the blandishments of Harvard or California. It is therefore good news that funds are being earmarked to boost the pay of a few dozen "world-class" scientists. These should mostly be channelled towards those here, rather than offering preferential deals to lure back those who have already left.

In the long run, the career patterns of now-young scientists will be even more crucial to the UK's scientific standing. This is a less tractable problem, since the numbers involved are far larger.

My Cambridge colleague Colin Humphreys is a world expert on the science of materials - a field crucial to all physics-based industry. He recently analysed the fate of the 27 excellent young researchers who had been in his group and who were still doing science, either in universities or in industry. Only eight were still in the UK. Most of us could provide equally depressing figures. This is sad, but not surprising. For a young PhD in physics, the starting salary of around £16,000, and the prospect of an insecure peripatetic life until their mid-30s, is a deterrent to all but the most narrowly dedicated. In consequence, we are jeopardising not only the UK's standing in key technologies, but the quality of the next generation of university teachers.

In the biomedical sciences, the situation has been somewhat alleviated by the Wellcome Trust, which raised its salary scale by 30 per cent. The physical sciences (the basis for the IT revolution) are more dependent on public funds. The Government should, as a start, enable the research councils to follow the Wellcome Trust's lead. But even that would be only a first step to what is needed to match the opportunities offered elsewhere.

Earlier this month, the UK hosted the International Physics Olympiad: teams of five 16- to 18-year-olds from 63 different countries spent a week in Leicester, solving challenging scientific problems. The results offered some insight into the contrasting educational styles in different countries. The best performances came from China, Russia, India and Hungary. We shouldn't read too much into this, because the teams for these countries probably received more intensive coaching. But nonetheless we should be mindful that, in science (as perhaps in tennis) the UK fails to tap and foster excellence.

The US team in the Physics Olympiad did well but not outstandingly - similar, in fact, to the UK. America has a poor high-school system, and fails to tap the talent of its own "minorities". Despite this, it ranks as number-one in most "cutting-edge" sciences. This is partly because of the competitive ethos and risk-taking culture that prevails there. But it owes a lot also to something else: the influx of foreign talent.

The most outstanding students from the Far East - even from India, where there were traditional special links with this country - now aspire to Harvard and Caltech rather than to Europe. For would-be computing entrepreneurs in Bangalore, there is a well-travelled route to Silicon Valley - not to the UK.

And the US gained a further recent influx from Eastern Europe. Russia's intellectual tradition in maths and physics was second to none, but it didn't survive the dissolution of the Soviet Union. The consequent exodus benefited primarily the US: many outstanding physics and mathematics professors now in US universities came from the former Soviet Union.

There is no law of nature that renders the US, to such an overwhelming extent, the destination of choice for mobile talent. We are, in most scientific fields, second only to the US. Moreover, the primacy of the English language gives us a head start over the rest of Europe. Our universities already exploit these advantages to attract overseas students. But at the level of very top talent, we are still losing out. It is good news that Government has said that it will change policy so that we no longer suffer the "own goal" set by immigration and work-permit restrictions.

When talent - especially cross-disciplinary talent - clusters together, the law of increasing returns applies. In all "priority" fields of science, the UK should aspire to having one research centre in the world's top five. This is a realistic goal, but requires concentration of resources, and enhanced recognition for outstanding people (junior as well as senior). Without such measures we'll surely end up with zero rather than one "world-class" research group - saving little money, but achieving far less.

We should exploit the UK's competitive advantages - a strong scientific base, and an English-speaking culture - and aim to be the preferred location for international research centres and consortia. Why shouldn't we set our sights high, aiming to expand our world-class science base, and achieve a "brain gain" as well?

Sir Martin Rees is the Astronomer Royal and Royal Society Research Professor at Cambridge University