My Lords, I agree with this proposal. We need an advanced project agency similar to ARPA. However, in setting up this agency, it is important that we understand what makes these agencies successful, and I think we are on the way.
To declare my interests, I worked for IBM in the USA for about 30 years, in its research and development laboratories and as a member of its corporate technical committee and science advisory committee. Additionally, and related to the US agencies, this year I chaired a sub-committee of the Draper prize committee of the US National Academy of Engineering. The Draper prize is the academy’s top prize. It has been awarded to those responsible for ARPANET, GPS and several other outstanding achievements of ARPA and DARPA over the years. The Queen Elizabeth prize for engineering has also been awarded to those responsible for the internet and GPS. I also declare that I drew together and chaired the first committee of judges for that prize.
Therefore, I have spent a lot of time studying how these remarkable accomplishments were realised and the characteristics of those responsible for their successes. As has been extensively discussed over the last year, ARPA and DARPA have contributed significantly to the dominance of the US in many high-technology industries, but of course they have not done these things on their own. They have drawn on industrial companies, other government agencies and universities, weaving together diverse capabilities to provide solutions to perceived needs. They did not invent these solutions, although many inventions emerged in developing them. Their genius was in pulling together the ingredients from the vast worldwide reservoir of science and technology. Their project leaders were noted for their breadth of expertise. They are a select group of highly talented individuals with exceptionally broad knowledge of science and engineering, and of the interfaces between the scientific disciplines—people who, for example, can tell whether a problem encountered in a highly complex computer-controlled system is a software or a hardware problem, or a matter of the science.
These exceptional people are paid a lot of money by UK standards. They are also obsessively focused on attaining the goals of the system that they are building and are not easily tempted to explore the new discoveries that invariably emerge when one builds new equipment. That is the regime of science, where the aim is to explore and extend human understanding. It is not the stuff of a project agency. In the USA, it is handled by the National Science Foundation. I have asked my friends in the US whether it would be a good idea to put their ARPA inside the National Science Foundation. They just laughed. To quote Dr Highnam,
the ARPA project manager and office director who spoke to the Commons Select Committee on Science and Technology:
“DARPA is not a blue-sky research place; we do not do that. Even with our fundamental research we know where it will be applied if we can make the science possible, all the way through to the higher technology systems programmes.”
They are not LMBs, which are temples or palaces of scientific genius, not project agencies. ARIA must select leaders who think like project agency managers and have this vast reservoir of knowledge. It is about project management and combining the knowledge and expertise that already exist, more than it is about invention, despite the name that has been given to this agency. I crossed out “inappropriate” but the noble Viscount, Lord Stansgate, sounded as though he would like to put it back in.
Successful high-technology projects need, as far as possible, to be free from time and money constraints. Therefore, the US agencies have been granted a lot of independence and freedom from continuous assessment —something that has rarely, if ever, been granted by the Treasury here. It is reassuring to see that ARIA is to have a minimum life of 10 years. This does not mean that it must be isolated. It will need to have close relations with Innovate UK, drawing from it the raw material of technological advancement and knowledge of where the skills to effectively apply what innovators have already extracted from the science reside. It must also have intimate knowledge of what is happening in industrial R&D laboratories and in universities. It will not be easy to be clear about the interface with Innovate UK, because Innovate UK was itself given many of the aims that have now also been given to ARIA.
The major advantage of forming this new agency is that it will not have to compete directly with the research councils for its funding, nor live within the regulatory structure of UKRI. Fortunately, there have been some very helpful recent changes in the management of Innovate UK, especially the appointment of Indro Mukerjee as its CEO, who understands project management. These changes should enable Innovate UK to play an effective role, working with ARIA, finally to provide competitive technology transfer in the UK.
However, I am still worried that we are at risk and will not learn from the past. After all, if Innovate UK had achieved what it was meant to—to drive technology transfer—we would not need ARIA. I was amazed to read that it was proposed by some that ARIA should be placed within UKRI, ensuring that history would repeat itself and ARIA would also fail by having to compete for funding using metrics designed for science rather than technology transfer. That is not to mention the regulatory structure of UKRI, which, while excellent for pure science, has not been optimum for Innovate UK.
Finally, how will the catapults, which were also meant to solve our technology transfer problem, fit into this confused cluster of councils and agencies? If there was more time—which there clearly is not—I would ask how it all fits with the grand challenges and the industrial strategy, but others have done that.
8.26 pm