Definitions of R & D : report with evidence.

  • Great Britain. Parliament. House of Lords. Science and Technology Committee.
Date:
1990
Catalogue details

Licence: Open Government Licence

Credit: Definitions of R & D : report with evidence. Source: Wellcome Collection.

    47/148 (page 45)
    2 March 1989] [Continued [Chairman contd.] measures. There are small firms which have technological activities which are called design and not R&D. So I think one is forced to explore other measures. One measure that we have used at SPRU is the counting of innovations, but it is very time- consuming and you cannot do it on a regular basis in Government. Another possibility would be to use patent statistics, but there the problem is identifying firms below 1,000 employees. We hope we will be able to do that in our research programme in the next five years. In the meantime, nobody has developed a regular review in any country which will reproduce retrospectively and reasonably economically the technological activities in small firms. Lord Nelson of Stafford 18. How do the Germans do it if they produce such good figures? (Professor Pavitt) Basically they do not do much better than we do. This is not a particularly British problem; it is a generic problem with R&D statistics. 19. So for comparison purposes it is probably not all that important, because they all have the same problem? (Professor Pavitt) Exactly, that is right. Where you might have a problem is in a country like Italy, which compares unfavourably with R&D statistics. We have a post-graduate student who looked at patenting statistics in Italy, and found that a lot is happening in small firms in and around textiles, leather and machinery—very Italian things like Benetton—that are very important in Italy with regard to technical change. Chairman - 20. Are there any other ways, other than by looking at patents, in which you could try to assess innovation? Output rather than input? (Professor Pavitt) 1 did not say I used patents as an output measure; I use them as a proxy measure of technological activity. There are reasons why it is dangerous to think patents as output, simply because patenting is used differently across industries. I do not know whether there are other methods. Lord Chorley 21. Are you talking about inputs or outputs? (Professor Pavitt) Activity. Lord Chorley: That is input, I think. Baroness Lockwood ~ 22. What is the difference? (Professor Pavitt) If I take, for example, the counting innovations as an output measure, you will find there is a great deal of incremental improvement which you do not capture by measuring outputs of specific innovations. What I mean is that one is measuring the complex phenomenon of technical change, and it is like putting instruments into a complex process. You can measure and monitor bits and pieces of it through R&D, patenting, through counting discreet measured innovations, through asking experts in other countries how good is this company or country, and getting judgments of peers, if you like. All these measurements are imperfect. What one is doing is putting them together to see if they tell a consistent story. Eventually one hopes to get productivity measures of all the technology inputs compared to all the technology outputs, but I do not think one is in that position yet. One is in a position of measuring imperfectly various parts of a complicated process. Chairman 23. But to go back to your comments on the inadequacy of industrial statistics, is that because the Department of Trade & Industry is not asking enough of the right questions, or questions in sufficient detail, or is it because of reluctance by industry to produce the information? (Professor Pavitt) | cannot answer that question. All I would say is, back to the late 1960s, much more detail and regular statistics were collected in what was then the Ministry of Technology. I-am not making a party political point, but at that time more resources were being given to this particular activity, and there did not seem to be any reluctance at that time by industry to supply quite detailed statistics. Lord Nelson of Stafford 24. It has been suggested that industry should be obliged to publish its figures on R&D, but it does not sound as though they are going to be of very much value, even if they are published, judging by the inadequacy of them? (Professor Pavitt) 1 think, for large firms quoted on the Stock Exchange, such data would be extremely valuable. In fact, there have been studies done in the United States, where R&D expenditures are subject to mandatory disclosure for quoted firms, which show that these data are taken into account by stock market analysts in establishing stock prices. 25. If the statistics are not worth anything, it could be very dangerous, could it not? (Professor Pavitt) 1 am one of those people who believe that one makes progress by developing statistics and improving them. If one believes in markets in the allocation of resources to technology, their efficient functioning depends on the availability of good information. We cannot expect markets to allocate resources efficiently if there is inadequate resource with regard to modern companies, which is their technology and R&D resources. It seems to me there is a very strong case, especially in a decentralised market system, for getting even better information available to a wide number of people, in order for it to work effectively. 26. In other words, you are saying this would be a move in our getting better statistics? (Professor Pavitt) 1am saying if you want markets to work in a decentralised and efficient way, one of the pre-conditions is having good information for decision-makers about the system.