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.

    50/148 (page 48)
    2 March 1989] [Continued [Chairman contd.] (Professor Pavitt) | thought you might ask me that! 46. Are you saying that either with present statistics or better statistics, if you were somebody with money in the market you could forecast that that is a good firm to put your money into now and that is not? Looked at from the Government’s point of view, is that something which the Government should support or not on the basis of the statistics that are made available? (Professor Pavitt) I think no statistics make decision-taking easy. There are no statistics on R&D or anything else that will tell you which horse to back, or which football team or which firm to put your money on. That is why people are paid sometimes quite high salaries to read the statistics and then to use their judgement to make decisions. I am often asked, “Why do you not tell policy-makers what they should do?” and my answer is, ““They are paid quite high salaries to decide what they should do. It is not for me to tell them.”” What I think the data are there for —and this is something you would understand — is intelligence. Intelligence is always imperfect; you get it from various sources and you have to cross- check. You want to know what your partners and enemies or potential enemies are doing, and you want all the systematic information available. You know it is imperfect, and it cannot predict the future, but you hope that by having it you will take better decisions than when not having the data. It does not take away the difficulties of interpretation by the policy-makers, but hopefully it makes them better informed. I am impressed— and I do have to be careful what I say —that the Japanese collect a lot of data and interpret a lot of data on R&D and other things. 47. The Japanese government? (Professor Pavitt) The Japanese government and Japanese firms. They publish quite a lot of data. Lord Chorley 48. So you would add one more point, that we ought to spend much more doing it? To be sufficiently useful it would be worth spending a lot more? (Professor Pavitt) | have an interest to declare in this, being an academic research institute in this area. It helps to make our job easier and more interesting to have good data. I would only point out that these data on R&D, like data on training, help concentrate minds on what makes advanced countries, high-wage countries and firms, competitive. If you compare the quality of the data we have on R&D and training with financial data it is grossly inadequate. I am also struck by the growth of specialised service firms like Derwent Patents, and big firms are increasingly interested in obtaining the data on what competitors and partners are doing in technology. There are entrepreneurs collecting not general public purpose Statistics, but private statistics and making money out of it, which suggests that there is a demand. 49. I seem to recall about two years ago the DTI launched something called’ the The Tradeable Information Initiative. I have not heard since whether that launch was successful, that is to say, whether anything happened or not, but I think the idea was that if there was value in this data people would buy it and the Government would be repaid for the effort. (Professor Pavitt) | was very careful to say where you have a clearly defined and _ specialised requirement for it. The point about R&D statistics is that interest in them is dispersed amongst a great many users. They are one input amongst others that will go into decision-making. They have the characteristics of a classic public good. They are of use to many but the incentive for any user to develop them is insufficient. Even Milton Friedman would say in such circumstances there is a strong economic case for public subsidy, and that is why governments subsidise the collection of statistics which have this general purpose function. In that sense, coming back to the point I made earlier in relation to Lord Nelson’s questions, for the efficient working of markets one needs better and more rigorous published information in this area. Chairman 50. In general you would be in favour of sticking as closely as we can to Frascati and not introducing new terms like “‘strategic”’ or “near market’””’ or anything else? (Professor Pavitt) That is right, yes. 51. You would not want to put in “strategic” as a subdivision of applied research? (Professor Pavitt) No. I think there are advantages to simplicity and continuity. The only area that I think does require a somewhat radical re-think is the sofware question which I raised. Lord Nelson of Stafford 52. In your paper you raise something not taken care of here. You say that statistics do not reflect technological activities undertaken in production engineering departments of large firms. That is quite big money? (Professor Pavitt) Yes. 53. How do you think it should be done? (Professor Pavitt) That is precisely the issue we talked about earlier in relation to aerospace: What comes after the R&D? I think you need special surveys. 54. Frascati does not deal with it very well? (Professor Pavitt) That is where patenting statistics are useful. You have a much better feel for the production technology from patenting than from the R&D statistics. That is another way of getting at the problem. There is a case for special surveys for post- R&D activities in certain industries. It is not important in pharmaceuticals; it would be a waste of time, but it would be useful in the automobile and aerospace industries. I know special surveys have been undertaken in Canada, the United States and Israel by academics, so you have data on post-R&D activities which turn out to be of the same order of magnitude as R&D itself. 55. If we look at Japanese statistics, their very successful activities over the last 20 years have been due to their devotion to production development? (Professor Pavitt) Yes.