Still room for improvement to ensure €70B Horizon 2020 delivers on innovation objectives

Improved support for technology start-ups and better systems for handling research outputs are needed to maximise returns on Horizon 2020, according to a new study looking into how the €70 billion research programme is shaping up.

Research Commissioner Máire Geoghegan-Quinn has personally pledged to reduce the bureaucracy that is synonymous with EU research programmes on more than one occasion, promising simpler reimbursement rules, simpler overall objectives and simpler forms.

However, more is needed if Horizon 2020 is to deliver on one of the central objectives, which is to involve more SMEs in carrying out R&D and get them acting as conduits for translating research and pushing innovation through to the market, says a new report, "Europe’s ‘Horizon 2020’ science funding programme: How is it shaping up?"

“Getting rid of the burdensome tendering process would really help to foster a link between small businesses and science,” says one of the authors Michael Galsworthy, Senior Research Associate at University College London. “Small companies waste scarce time compiling bids which fail, and investigators are bound to strangers who present the lowest bid regardless of passion or competence,” Galworthy said.

The Commission claims it is listening. "The article puts forward some very interesting ideas, some of which are already part of the discussions for Horizon 2020," Michael Jennings, Geoghegan-Quinn’s spokesman told Science|Business.

Proper use of public money

However, while the paper calls on the Commission to remove compulsory tendering for sub-contracting where the sums fall below the procurement thresholds for public sector authorities, a Commission official has said this is not feasible given the responsibility the EU carries for the proper use of public money.

Galsworthy claims suitable safeguards are already in place. “All projects and budgets must be approved by the Commission. This is a check that the amount of money seems sensible for the job done,” he said. In addition, Horizon 2020 will require that projects bring together three partners from three different countries, meaning there is an international team to bear witness to the principal investigator's ethical behavior. Everyone has something at stake in terms of the reputation and success of a project output, so it makes no sense to choose an under-performing partner, Galsworthy noted, adding, “It you really wanted to be corrupt, why wouldn't you write your friend into the project from the beginning? 

Information black holes

There is another major shortcoming with the Horizon 2020 programme as it stands, and that is around the requirements and systems for handling research outputs.

The US National Institutes of Health (NIH) allows anyone to freely search its database of research projects by geography, year or subject area, and to access resulting papers and patents. But there was no such tool available to Galsworthy and his fellow researcher Martin McKee, Professor of European Public Health at London School of Hygiene and Tropical Medicine, when the two were working on an EU-funded project, ‘Health Research for Europe’.

When trying to compile a record of all health-related research funded by the EU's Fifth and Sixth Framework Programmes, they found the EU's Community Research and Development Information Service (CORDIS) to be of little use, claiming the experience was like looking into a black hole. In order to assemble the record they had to manually classify over 4,700 projects.

Aside from the issue of allowing open access to outputs of publicly-funded projects, mapping of research is important in ensuring funding matches need, for example by cross-referencing with statistics on disease burden or patient demand. A detailed online hierarchical categorisation of EU investments would allow national funders to see collaboration opportunities or gaps.

As one stark example of how this might drive R&D funding policy, statistics show that the original 15 EU Member States have received 34 times more health research funding under FP7 than the 12 newest members. McKee and Galsworthy conclude that even allowing for gross domestic product and population differences, this represents dramatic underfunding.

Access to publicly-funded research

The tools are on hand to provide comprehensive and reliable access to research funded by the EU, for example the Open Researcher and Contributor ID (ORCID) database, which reliably attributes research outputs to their author by assigning every scientist a unique digital identifier.

Galsworthy suggests the EU could require all project applicants to sign up to ORCID. "This would make the selection procedure much more efficient," he said. "The ORCID number would allow the EU to judge the suitability of potential participants by giving them access to their previous work. Once the EU has decided to fund a project, they can then use the service to track the researchers' outputs."

Tracking could be further improved by assigning each project a code to be cited in papers. Entering a US grant code into PubMed, the free search engine of the US National Library of Medicine, allows all outputs from that project to be retrieved almost instantaneously. Galsworthy and McKee call for the EU to join other major funders in signing a contract with PubMed.

The Commission has stated that all articles produced with funding from Horizon 2020 will have to be accessible. Articles will either be made immediately accessible online by the publisher, so called Gold open access, or researchers will make their articles available no later than six months after publication in Green open access.

The promises on open data now need to be translated Galsworthy said. "Lots of research funding bodies are talking about data sharing but bold declarations are nothing without real action.”

http://hsr.sagepub.com/content/early/2013/04/30/1355819613476017.abstract