CORDIS Archive

March 2001

INNOVATION INDICATORS

Figuring it Out

Our current understanding of what causes innovation is far from adequate. Perhaps this is unsurprising - it is a relatively recent arrival in the economist's lexicon, and as the Commission's Jean-Noël Durvy remarked, the innovation industry does not have at its disposal the kind of resources that are spent on agricultural statistics. But what should we measure - and how?

Eight workshops examined different aspects of innovation measurement.(1) How to tackle the service sector? How innovative are traditional manufacturers? How to get a handle on inter-firm collaboration? What is the impact of outsourcing, employee involvement, financial incentives, technical standards and factory location? How to enable regional administrations to free up innovation potential? What can we learn from scoreboards?

Old ideas of what's new

It was Professor Keith Smith of the Oslo-based STEP Group who set out the parameters. Our current data are all in one way or another flawed approximations. Neither high R&D spending, nor a high rate of scientific publications, leads automatically to innovation, and not all innovations are patented. Nor should we confuse innovation with new technology. The more traditional sectors of the economy, such as food processing, spend little on research directly, but make use of complex knowledge inputs and are continuously innovating on a very large scale.

Ironically, the central problem is that the definition of innovation has not kept up with the times. The figures that everyone works with focus primarily on manufacturing industry, and omit what is now the greater part of our economies - the service sector. They look at new products rather than organisational change. We should not make the mistake of imagining that the world consists only of what we can measure.

Indicators, as Professor Lena Tsipouri of Athens university said, need to be robust, feasible and relevant. They are robust if their value cannot be contested, feasible if the data are available at reasonable cost, and relevant if they bear a direct relationship to policy objectives. Finding indicators that meet these three conditions is not easy. "There is often a trade-off, and the choice of indicators is a political question," she said. "The technical tools are there - what is needed is to give the politicians the mandate to apply them."

But there are also more practical difficulties, including that of data incompatibility. The quality of the data improved markedly between the first Community Innovation Survey (CIS1) in 1993 and CIS2 in 1997, but is still far from ideal. Furthermore, data needs to be gathered more frequently than once every four years. And though researchers would dearly love to have access to low-level data, confidentiality considerations often prevent this.

High-tech - no panacea for jobs

Deeper analysis seems to belie the common assertion that knowledge-intensive industries are the motors of growth. Figures quoted by Professor Smith show that between 1980 and 1995 in the United States the share of high-tech in manufacturing output grew from 10.5% to 15.8%. But what is often overlooked is that over the same period manufacturing itself shrank from 21.5% to 18.5% of the economy. The net growth in the high-tech sector is less than 1% of GDP. What is more, in 11 of 15 OECD countries surveyed, high-tech employment is actually falling. So to pin hopes of near-full employment on the high-tech industries would be a serious mistake. We have to look for innovation everywhere.

The medium-term intellectual challenge is to identify the systems that support growth. What are the complex knowledge bases at firm, industry and economy levels? How does knowledge flow between industries? But time is short. Pragmatically, we must look at what works - survey what everyone is doing, and emulate the best we can find. This benchmarking technique was recommended by the Lisbon summit a year ago, and has already been put into practice in the form of the European Innovation Scoreboard(2).

Better decisions across the board

The importance of innovation must also be brought to the attention of new audiences, so that improved information is applied in practice to result in improved policy-making. Europe's business association UNICE has taken a welcome step not just in conducting a wide-ranging survey of innovation from the employer's perspective, but in marketing it energetically. Their third benchmarking report Stimulating Creativity and Innovation in Europe was launched at the high-profile European Business Summit in Brussels last June(3). But the link between science and policy-making remains as tenuous as ever. There is still a lot of room for debate as to whether policy measures such as tax reductions will stimulate innovation.

If we are not to lose our way as we move into the knowledge-driven economy, we need milestones by the roadside. Large regional aid budgets are at stake. Eurostat is working on new indicators such as 'employment in knowledge-intensive sectors', which are intended to complement the results of the forthcoming third Community Innovation Survey (CIS3). In parallel, UNICE intends to benchmark the 'new economy' this year. The job requires not just scientific precision but the imagination and courage to dare to measure the immeasurable. As Rémi Barré, Director of the Observatoire des Sciences et Techniques in France summed up: "Like biotechnologists, innovation theorists must learn to live with controversy - to realise that they are useful to society even though they know next to nothing."

(1) The papers presented are at
/innovation-smes/src/statconf5.htm
(2) See the special edition of November 2000. Further information is also available at /trendchart/
(3) See 'Freedom with Responsibility', edition 5/00.

Contact

• K. Schwall, European Commission
  Enterprise DG
  Innovation Directorate
  Innovation policy
  Tl. +352 4301 3380
  9Fx. +352 4301 34129
  klaus.schwall@ec.europa.eu
  /innovation-smes/src/statconf.htm