The European Union is run by its council of ministers and its parliament, with the European Commission as its executive body. But there is another component that is generally unloved and yet has a crucial role: the Joint Research Centre (JRC). The seven large institutes that make up the JRC have remits that range from energy and environment to health and security. With a budget that approaches €400 million (US$520 million) per year, the JRC is responsible for providing the scientific and technical support for EU policy. And according to a fairly positive evaluation released last week, led by David King, former science adviser to the British government, it now does this quite well.

The dismal reputation of the JRC originated in 1980s, after changing priorities in Europe had forced the centre to shift from its original mandate of researching nuclear energy, safety and security. It reinvented itself, and diversified its research fields, in a largely undirected and unmonitored way. Politicians soon started to complain about its lack of a clear mission and its inefficiency. In 1998, the institutes' labs were relaunched with a new, customer-orientated mission, tightly harnessing them to EU policy support and fulfilment.

The various JRC centres are now reasonably efficient machines, pretty well stripped of the dead wood that had been dragging them down. They fulfil just about every task their customers — that is, the EU bodies — set them, often with great success. The King report acknowledges, for example, that the biotechnology division has become the world leader in setting standards for detecting and monitoring genetically modified organisms in agriculture and food, while also playing a key part in the implementation of the controversial new Registration, Evaluation, Authorisation and Restriction of Chemical substances (REACH) regulations.

In the process of becoming customer orientated, however, the JRC lost its independent research activities. Does this matter, given that they were not highly regarded anyway?

Yes, it matters. The role of science in policy has become ever more important. Crucial decisions about climate, energy and the like rely on the highest quality of scientific information. JRC scientists need to be able to interact intelligently with cutting-edge research in the academic community, and to extend it appropriately.

But high-level scientists need an appropriate intellectual environment to work in, and this is where the JRC has had its hands tied. Right now, JRC researchers are running to stand still. As new tasks stream in from their demanding customers, research projects enthusiastically started during a previous task are dropped — even if they might have had long-term value for policy. The scientists tend not to publish much in academic journals, as this has not been considered one of their 'deliverables'. Most JRC science is written up in internal reports. These are not ideal conditions for attracting and keeping top, ambitious scientists.

This is why the King report's proposal that a small proportion of the budget be directed into exploratory work is welcome, as are two other radical proposals, to some extent related.

First, the JRC needs to develop its own long-term strategies so that it can plan more rationally. Although it has 2,750 staff, the scatter-gun tasks required of it spreads them too thinly. So the strategies should also set out criteria for which tasks to accept so that critical mass can be achieved from limited human resources.

Second, the JRC needs more political responsibility. Given its accumulated knowledge and experience, and its close contact with the scientific community, it is well placed to see problems coming. Structures should be set up to allow it to advise, not just serve, its customers.

In short, the JRC needs to be allowed to grow up. Europe's council, parliament and commission now need to shed their distrust and make the JRC an even more useful institution than it has already become.