PUBLIC LECTURE
Putting our science to work
Insights into the commercialisation of science
The Shine Dome, Canberra, 13 March 2007
Thomas M. Connelly Jr. (external link)
Executive Vice President and Chief Innovation Officer
DuPont Electronic and Communications Technologies, USA
Tom Connelly reviewed DuPont's growth strategies and how it is adding value to research undertaken in universities and other research institutions by assisting researchers to commercialise
their research.
(Click on image for a larger version)
Tom Connelly: Thanks, Kurt, for those kind words of introduction. I welcome this opportunity to talk about at least one company's experience in bringing technology to the marketplace.
(Click on image for a larger version)
I will start with a few key messages.
It is clear to anybody who spends his time in industry that the world is becoming more complex. It is certainly becoming more competitive, and competition is coming from new places - places that we didn't look to as competitors, even 10 years ago.
The research process is truly a global one and hence not only does competition exist in many new places, but opportunities for collaboration do as well.
Professor Henry Chesbrough, formerly from Harvard Business School and now at University of California Berkeley, coined the phrase and wrote the book called Open Innovation. It is something that we at DuPont practise and that I think all leading companies at this point are practising. I will certainly spend some time today talking about how to do that.
(Click on image for a larger version)
If you will permit me a little bit of time to explain the background of the DuPont Company, I think it might be helpful to the later presentation. I will try to get through this quickly.
DuPont is a science company. In the somewhat aspirational words, we position DuPont as 'the world's most dynamic science company'. We make the point very clearly that we can't leave our science in the laboratory; we need to move it into the marketplace, making people's lives better, safer and healthier. And many of the thrusts of our innovation are directed toward those endpoints.
(Click on image for a larger version)
As we talk about DuPont as a science company, it is really the depth of the science that we try to emphasise. DuPont has been around for 200 years; it is one of the older industrial enterprises in the world. We spent our first hundred years as an explosives company, starting with black powder and moving into high explosives - dynamite et cetera - throughout the 19th century. During the 20th century we moved into chemicals and plastics, and had a brief foray into energy with an oil company, Conoco.
As we dared to think about preparing the company for our third century, we realised that what had brought success in the 20th century was necessary for our future but it wasn't sufficient for our future. Hence our desire to find more areas in which to innovate and find more areas where our company could create value, and hence the emphasis on building on our tradition and heritage in chemistry and materials areas, and adding to that the capabilities in biotechnology - specifically in ag biotech and in industrial biotech areas - and also working in the areas of electronics and electronic materials.
(Click on image for a larger version)
As a nearly $30 billion dollar company we organise our businesses into five 'growth platforms', each of the order of $5 billion. They are DuPont Safety & Protection, Electronic & Communication Technologies - one of my areas of responsibility - Coatings & Color Technologies, Performance Materials, and Agriculture & Nutrition. You can see our growth strategies listed below those, and certainly I spend my time as Chief Innovation Officer very much focused on putting our science to work.
(Click on image for a larger version)
As a technology organisation, we have 4400 scientists and engineers in technology roles. We spend $1.7 billion on technology, of which $1.4 billion is R&D effort and the remainder is technology work associated with our manufacture or with our marketing activities.
We have 25 R&D centres worldwide, and the next slide gives a sense of how that has moved globally.
(Click on image for a larger version)
Shown at the lower left is the DuPont Experimental Station. We have been doing research at that location for over 100 years - to say 'since 1803' is a bit of an exaggeration for our research effort, although the first US patent granted to DuPont was to our founder, the French émigré Eleuthère Irénée du Pont, who came to the US, started the company and was granted our first patent. So at least we were involved in technology work from the earliest days.
Our Experimental Station is the site of many of our most important innovations and it also represents perhaps a third of our total technology work.
You can see that we have moved globally, first into Canada, where our Kingston centre has been operating for 50 years. It is the site for many of our process innovations in chemical engineering. The Meyrin centre, in Switzerland, was opened in the late 1980s. I had the privilege of being the second director of that laboratory. Wuppertal, in Germany, is the centre for our coatings technologies, which arrived in DuPont via acquisition in 1999.
As you can see, our recent laboratories have been established in the Asia Pacific region. In early 2005 we opened our first major research centre in Shanghai, China - in the Zhangjiang High-Technology Park, for any of you who are familiar with that region. We have about 200 researchers there and we are already in phase 2 of expansion of that facility. That same year we opened a small research presence in Korea to meet the needs of our electronics customers there, and the following year, 2006, we opened in Hsinchu, in the northern part of Taiwan, again specifically dedicated to our customers in the electronics industry. Just last month we announced our intent to build a major knowledge centre, as we are calling it, in Hyderabad, India. The focus of this will be ag biotech, initially; research in other areas; technical support to our customers in India; and also a centre for our engineering activities in some of our information technology activities. In total it will be a large centre, housing 500 to 1000 people. So we continue to expand globally.
I must point out that in Australia we have a small technical centre founded by Mr Leo Hyde, who is sitting in the back of the room this evening. It provides very important technical support to our customers here in Australia.
(Click on image for a larger version)
With that background, let me switch gears and talk a little bit about DuPont's approach to collaborative education. I have explained our long history in terms of our company, I have talked about our long history in terms of R&D. We also have a relatively long history in terms of our involvement in external research.
The earliest example I could find of a development of commercial significance dates back to 1925 and a professor of chemistry, Father Julius Nieuwland, of the University of Notre Dame, in the United States. In the early 1920s, Professor Nieuwland presented a paper at the American Chemical Society meeting on work he was doing with acetylene chemistry and the production of monovinylacetylene, which DuPont was interested in for the synthesis of chloroprene for the production of neoprene, our first commercially successful synthetic rubber. It was actually Professor Nieuwland's chemistry that was used in the commercialisation of neoprene and that technology was practised from the 1930s right through to the 1960s.
In the 1920s the University was not prepared to accept direct monetary compensation. Most universities have overcome that initial objection and we don't have that as a barrier today. But we did maintain very good relationships with the University and Professor Nieuwland, and in fact provided support in other ways through providing laboratory equipment, subscriptions to journals and other things we could do there to further chemistry education at the University of Notre Dame - even though this was done literally as a gentlemen's agreement.
We recognised at that time the need to do more and more in terms of our relations with universities, but I must say most of our work at that time was really directed toward developing relationships for recruiting purposes, so we could access the top science talent from the universities that we were working with at the time.
(Click on image for a larger version)
The list on this slide shows that fully half of DuPont's technologies and products had their origin outside the company. We accessed technologies for products such as cellophane and rayon in the 1920s and '30s, and then later for polymers such as polyethylene or polyethylene terephthalate polyester, if you will. Those are typical of the products where the DuPont technology came from outside the company.
Amongst the products for which we developed the technology ourselves were nylon, Teflon, Freon, Lycra, Tyvek, Kapton, and later our sulfonylurea herbicides. All of these were tremendously successful products where DuPont invented the product, invented the category of product, was first to market and established some really iconic brands associated with those businesses. And that became the way we did research from the 1930s into the 1950s, '60s and '70s.
Another thing that is important to me in all of these examples is that this wasn't a question of collaborative research to develop the technology; it was a question of literally accessing what were essentially complete technology packages from the outside.
(Click on image for a larger version)
Through the years there have been a number of important milestones, in the United States at least, that define the way universities and industry work together. I will go way back to 1862, to the Morrill Act, which established the so-called land-grant universities within the US. This was an act of the US Congress that defined the role of universities in research of an applied nature. A lot of this initially was in areas of agriculture but it really defined a mission for a group of US universities in terms of bringing into practice science in a way that benefited society and industry.
By the end of the First World War, DuPont was active in supporting universities - most of this in the form of grants to universities in hopes of developing close relationships for recruiting purposes. By the end of the Second World War, research partnerships had sprung up between DuPont and a number of universities.
Then in 1980 the Bayh-Dole Act was another important and defining moment in the way universities worked with industry, because the Act allowed public universities in the US to maintain some of the gain they had from the research they had done. So they were able to license technology, collect royalties and keep those funds for purposes of the university, and share some of the benefit with the faculty associated with generating that technology. This spawned a tremendous amount of interest in US universities in revitalising their approach to research and turning that research into something of benefit to the university and the faculty.
There are similar acts that govern the US national laboratories - as I have shown here, in 1986 and '89, a kind of Bayh-Dole equivalent for the national laboratories was established. The Act encouraged the national laboratories to move beyond their national mission, in terms of developing technology that could be licensed and commercialised.
In response to some of these trends, changes occurred within DuPont's approach. I mentioned our relationship with the universities. Originally we had a program we called Aid to Education, which transformed itself into the DuPont Center for Collaborative Research and Education. When tech transfer offices started to spring up in the universities there were actually people who knew how to write contracts, cash cheques and do the kinds of things that were needed for these purposes.
The final point I would make there is that universities began to become quite important incubators of technology. There had always been a sort of incubation environment set up around universities. Around Boston the region called Route 128 was filled with small companies started, in some cases, by faculty members from MIT and other universities. Silicon Valley, in California, was the result of those very strong research universities in that area. But it happened without, in many cases, the direct involvement of the universities; it was more of an entrepreneurial spirit that grew up around the universities.
I had an experience recently where I met a faculty member from a leading research university who handed me two business cards: one was his card as a faculty member in the Department of Materials Science and the other was his business card as general partner for a venture capital fund. I looked at the two cards and said I thought both of these were full-time jobs. But apparently people now manage to live in both of those worlds.
(Click on image for a larger version)
What is Open Innovation all about? It is the recognition that companies can no longer develop all the technology that they need by themselves, in their own laboratories, and it is further recognition that no company can commercialise all the technology it has, without collaborations with others. So, both in technology development and in the exploitation of technology, companies recognise their need to have partners.
Further, companies need to define the mix of internal research versus external research that they want to pursue - decide what they are going to create internally and decide what they are going to source externally. And I would say, in crude terms, that this is almost a 'make versus buy' decision on technology. Before we launch our internal research program, we are going to have to ask the question: What can we get on the outside, how much will it cost us and how much time would it save us if we were able to access technology from the outside?
There are a few examples where this is going on to a very great extent. The Big Pharma model for research is running into these issues - there are fewer products in the pipeline and many of the pharma companies are very active in terms of trying to source molecules from the outside, going to start-up pharmaceutical companies. And they are positioning themselves more as the developers and the marketers of drugs, rather than the discoverers of molecules. One American company, Procter & Gamble, talks about replacing their R&D effort with a C&D - connect and develop - effort, where the 'R' piece of their R&D is sourced either internally or externally and then they take it from there and move to the marketplace.
(Click on image for a larger version)
This chart tries to pictorially represent what we are talking about - the idea of combining DuPont's internal capabilities in our large regional technology centres with the Open Innovation models, connecting with global corporations. These could be large-cap companies, start-up companies, national laboratories or universities, taking ideas wherever we want to and combining both our internal and external sources with our market knowledge as we try to develop those offerings and bring them to the marketplace.
I would like now to illustrate a few examples of what DuPont is doing in terms of those types of external collaborations.
(Click on image for a larger version)
My first example is work we did with the University of North Carolina. Professor Joe DeSimone developed technology around supercritical carbon dioxide as a process aid for conducting a wide range of chemistry, including polymerisation of fluoromonomers. (I would point out that Professor DeSimone has also launched at least two start-up companies, to my knowledge, so he is in this class of new academic entrepreneurs.)
We licensed Professor DeSimone's technology; we used it, scaled up to a commercial scale and are producing certain grades of fluoropolymers at our Fayatteville, North Carolina, site using that technology. Professor DeSimone's funding is largely through an NSF centre for green chemistries, in the application of this supercritical carbon dioxide to come up with more environmentally acceptable processes for production of materials.
But I would also point out that our relationship with Professor DeSimone is not our only one at the University of North Carolina. We also have had a successful collaboration with Professor Maurice Brookhart on late transition metal catalysts for polymerisation of olefins.
I think those results are a pattern that we have seen established - once we have found a university with whom we can work very easily, there is a tendency to go back and look for further projects that we can do with that university.
(Click on image for a larger version)
I have picked a couple of examples of partnerships in Australia. There was a very intriguing report, back in 2004, that was widely reported in the US - C&E News [Chemical & Engineering News] and other American Chemical Society publications - about the identification and isolation of a 'smoke compound', as it was called. This material, at very, very low concentrations of one part per trillion or below, was very active in promoting the germination of seeds. Being in the seed business, we were excited by the possibilities and immediately established contact. In fact, Hutch Ranck, here in the audience, and Leo Hyde were two of the people who approached the University of Western Australia and other institutes who were involved in this technology to set up a collaborative relationship, to understand and, it was hoped, to exploit the potential of this smoke compound.
I would say that to date this program has not moved forward toward commercialisation. There were a couple of issues that we faced and may or may not be able to overcome.
The first of those was that this material was very effective in germinating weeds. This came from the observation that after a forest fire, in a very short time new plants spring to life. How is it that there seems to be such an abundance of germination of seeds soon after a forest fire? So it was on that basis that the compound was isolated. But we have found that it was more effective in germinating weeds than it was for the row crops that we were interested in germinating.
Also, it was not surprising, for a material that is biologically active at such low concentrations, that there are some toxic effects associated with the compound that needed to be understood and worked through. But despite the fact that this particular program has not, perhaps, moved in a very direct way toward commercialisation, it did allow us to establish some important relationships that we value and where we continue to explore other possibilities for collaboration.
(Click on image for a larger version)
The second Australian example I want to mention is DuPont's long-standing relationship, going back 20-odd years, with CSIRO. It started in the area of agricultural chemicals and it has moved into a number of polymer subjects. I was at CSIRO earlier this afternoon and we will have a group of CSIRO visitors next week and then another group in May. This has been going on, on a regular basis, for some time. We have had some commercial successes, we have also had some programs that didn't move toward commercialisation, but that is the nature of research, as you would all certainly know.
Once again we see this collaboration starting in one area and moving into a very broad area with things ranging from polymer science and industrial biotechnology to agricultural subjects and electronic materials. So once again we have found a very broad basis for collaboration with CSIRO, starting from one particular subject.
(Click on image for a larger version)
My next example is one of DuPont's leading examples of Open Innovation. That is the DuPont MIT Alliance, or DMA as we call it. This work started back in the year 2000 when our Chairman, Mr Holliday, decided that we needed to accelerate our entrance into certain areas of biotechnology and industrial biotechnology. Dr Charles Vest, who is a member of our board of directors, was President of MIT, and through the vision of Dr Vest and Mr Holliday the decision was made to move ahead with the DuPont MIT Alliance; using MIT's science capability and DuPont's market insight and ability to commercialise.
We launched in 2001 in biotechnology; in 2006 we renewed this program for another five years. The total expenditure over that 10-year period will be of the order of $60 million. And as we renewed for the second five years there were so many DuPont scientists interested that we broadened the scope of it beyond biotechnology to include electronic materials, certain areas of nanotechnologies and specialised materials. So I think the pattern was well established.
Although research occurs in the areas of interest to DuPont, there is a significant educational component to the alliance as well. The educational component comprises two pieces. MIT had a need for fellowships for first year graduate students before they were associated with their projects; and so DuPont sponsors 15 to 20 first year graduate students as DuPont Presidential Fellows. MIT also participates in the education component by providing short courses for DuPont scientists and there is regular exchange in these areas.
Another important component is our intellectual property agreements, which we needed to hammer out as we put the alliance together.
(Click on image for a larger version)
So what are some of what we consider the unique attributes of the DuPont MIT Alliance?
The first is the breadth of MIT participation. We have had faculty members from more than 20 departments or programs with MIT apply for and receive funding under the DMA alliance.
A second aspect is that DuPont scientists or MIT faculty could propose projects for this, and in fact we have found over time that they tend to be jointly proposed now. So MIT principal investigators and DuPont scientists come together and really hammer out the programs they want to work on and jointly propose the programs and then the programs are jointly led. One of the things that Bob Brown, who was Provost at MIT and is now President at Boston University, said is that one of the distinguishing characteristics about this industrial research program versus some others is that it isn't a question of throwing the money over the wall to the Institute - DuPont scientists and their MIT counterparts are literally in weekly contact on these programs.
In addition to that project by project oversight, Bob Brown and myself, constituted the steering team. We met twice a year to review portfolio and progress. In addition to our two meetings as the steering team with our staffs, we had two events during the year, one at MIT and one at DuPont, where the scientists actually presented their work. So we would go up every autumn - as we still do - and let the graduate students present the work at MIT, and then every spring the faculty of MIT come down and provide the updates to us at our location.
Intellectual property rights have been an important aspect of this. We have special terms, which we haven't disclosed, that were received from MIT. We had one issue in that regard: the MIT Media Lab had their own collaborators, they had their own standard terms. They did not fit with the terms that we agreed to and the Media Lab weren't willing to change their approach, so we agreed to disagree and we have not funded any programs in the Media.
I have talked about the educational component.
One interesting aspect was the involvement of the MIT Sloan School, their business management school, who have not only helped us to develop business plans around some of the projects but have also helped in studies of our approach to innovation and looking at innovation in general.
(Click on image for a larger version)
I will mention just a few others. As a consequence of the close working relationship we have developed with MIT, when MIT prepared its proposal for a US Department of Defense sponsored program called the Institute for Soldier Nanotechnologies it decided from the earliest stage to involve DuPont in helping to write the proposal. This is a 10-year program at, I believe, $10 million a year for 10 years. MIT is the centre of this. DuPont was not a stranger to the areas of soldier technologies; through our work with Kevlar, ballistics, Nomex and flame-resistant uniforms et cetera we had a familiarity with some of the challenges in working on subjects such as chem/bio protection and active armour systems for ground troops. I believe that our involvement in this program was a direct result of the relationships we had developed with MIT in the previous program.
(Click on image for a larger version)
A second example of a program that brings together university, industry and also government is the Integrated Corn-Based Bio-Refinery (ICBR) program that we have joined. This is a program whose goal is to convert cellulosic materials to simple sugars for fermentation to produce fuel ethanol.
Our work in this area began in 2002. At that point the US Department of Energy (DoE) was interested in proposals in this area. They received 125 proposals. Our understanding is that we came first out of those 125. We received funding as a cost-share of $19 million from DuPont and $19 million from the DoE over a four-year period. None of the DoE money stuck with DuPont - we paid our own costs, the money went to our collaboration partners. Four or five years on, of those five projects initially funded by DoE, there is only one that is continuing to deliver results and that is the one that we have been involved in.
We got further confirmation of the value of this program when just two weeks ago the Department of Energy was interested in funding this technology, bringing it toward commercialisation, and our partner for commercialisation, the Broin Companies, received an $80 million grant - again a matching grant; they will put up $80 million of their own resources. DuPont will provide our cellulosic technology for a demonstration-scale plant of our plans for an integrated corn-based bio-refinery.
You can see that this consortium brought together the Pioneer unit of DuPont; John Deere for collection of the corn stover; the National Renewable Energy Laboratory, the government laboratory; Diversa, a start-up company for enzyme evolution, located in San Diego; and Michigan State University. Professor Bruce Dale, at Michigan State University, did the life cycle analysis associated with this approach. So once again this is a complex but very successful collaboration, involving a lot of partners.
(Click on image for a larger version)
I will begin to draw my comments to a close here with some of our lessons, learned from our work with collaborative research.
We need to start out with a clear idea of what we want to do and where we need to reach out and connect with people. I think the relationships that we build tend to transcend any single program. Any single program can be successful or not, but we find that places like MIT, the University of North Carolina, CSIRO and certainly, close to our headquarters, the University of Delaware, have been partners again and again for programs.
I would say that patents are always an issue, but we certainly have ways of overcoming those difficulties and this should not be an impediment to moving ahead with collaborative research.
There are some things that can be barriers to us in getting this job and negotiations can bog down in our discussions with universities. It can happen in our discussions with companies as well, and you learn very quickly the companies that are easy to work with, or the universities that are easy to work with, and those that are not, and simply choose your partners accordingly.
(Click on image for a larger version)
There can be issues in universities about becoming too close to industries and compromising the role, the mission of the university. There was an example a number of years ago in the US where major pharmaceutical companies made a large financial commitment to a university and the word came out that they were 'buying' the chemistry department in that university. So I think there can be backlash if these kinds of collaborative things are not done in the proper way.
From an industry standpoint we need to do this as truly collaborative research and not some sort of outsourcing mechanism. We have to recognise that the university's primary role is around education and research and there are students involved in these things. So when we start programs and we stop programs, we have to be respectful of the fact that we have students who are working on degrees and not leave students high and dry in the middle of their projects.
But once again these are not things that cannot be dealt with if the industrial partner approaches it in the right spirit.
(Click on image for a larger version)
We have been successful with a range of collaborations. From an industry standpoint, done well these can definitely provide a faster route to developing certain technologies. They provide a lower-cost route to take on a more speculative approach to research and, as I say, we have set these relationships up with other companies, with government laboratories, with universities at the institute level or the departmental level or even at the individual researcher level.
(Click on image for a larger version)
From a university standpoint, this does provide a commercial outlet and a source of income. It tends to share in work that is done within the university. And what we also find in our experience of very successful collaborations is the spawning of further ideas. In fact, many of our research partners, when they have an idea, will actually contact us and say: 'We're working on this project. We don't see a way to get it commercialised. Would you come in and help us with this?'
When we were looking at our DuPont MIT Alliance, Bob Brown and I were talking one day and reviewing various proposals that came to us, and Bob said to me that one of his faculty members had been complaining to him that the DuPont MIT Alliance felt more like DARPA than the NSF. And at that point we looked at each other and said: 'Well, we've got this just about right'. The intent is to develop technology and to get it into the marketplace, but we have to do it in a way that respects the role of the university.
(Click on image for a larger version)
Finally, partnership is the way to go. Industry needs to do more of this, not less of it. We have to do it in a way that makes sense for ourselves, makes sense for our partners. And choosing partners with complementary objectives for the research is the best approach. Building relationships at multiple levels with the institutes - at the individual researcher level, at the leadership level as well - is very important. And continuity is important in the relationship: the best things come out of a prolonged relationship. With the DuPont MIT Alliance, we initially launched it for five years but five years into it we recognised that a lot of the fruits of what we had started were going to come during the second five years and it was absolutely essential that we continue that.
I had the opportunity to talk to the presidents of a number of research universities a little while ago and I concluded my comments with the observation that at the end of the day we in industry have to recognise that what we offer to the universities they can receive from a number of capable industrial research partners and at the same time the research universities have to recognise that there is no shortage of very capable research universities. Neither what we have to offer nor what the universities have to offer is in short supply, so what we need to do is to go about it in a collaborative and mutually supportive way, and if we do that right, some very good things can come of the collaboration.
(Click on image for a larger version)
I will end there, and I thank you once again for your attention. If there are any questions, I'd be happy to take them. Thank you.
Discussion
Kurt Lambeck: Thanks very much, Tom, for sharing your insight with us. I am particularly struck by the approach you develop of picking an institution and then developing the partnerships in the recognition that these are very long-term ones. This is quite an enlightened approach to things.
Question 1: I can see the great advantage to a big company like yours of mining the intellectual capital of the universities for good ideas, but I am not sure that enough attention is being maintained, certainly here in Australia, in actually investment in that by the taxpayer, as it were, so that it is continued. I think you will just run out of good ideas.
You mentioned various down sides. You mentioned, for example, somebody you met who had two cards m he offered you his business cards, one with the company and one without. The Deputy Vice-Chancellor of this university has exactly the same problem, in that he has companies as well as being in charge of all the intellectual property of the university.
So my question is this. Some years ago I heard that Japan realised the problems and some universities in Japan decided not to go that way. I couldn't remember back to your slide as to whether DuPont had got into Japan and whether or not they had found a different way of behaving there.
Tom Connelly: It is a very good question, in that the role of the university in Japan was always quite different. It was very much the idea of supplying talent to industry but really a very hands-off sort of approach to that, in terms of actually working with industry. That situation has changed dramatically in the last few years, and in fact I have gone into Japan twice to speak on the subject of university industry collaboration in the past three years. I do know they are fascinated by models, they are concerned about stimulating innovation, and universities like Kyoto and Tokyo and the Tokyo Institute of Technology also, have really tried to reposition themselves opposite this.
I am not sure I would wish to see all the aspects of what is going on in the US in terms of the dynamics there, with universities open for business in a very direct way - I am not sure that is the ideal - but I think Japan is definitely taking a step in a new direction.
But, to your point: we have no research relationships with Japan's universities. I think that is something that is going to change over the next few years.
Question 2: I have two questions. One is about trying to forecast where you think the steady state will be for companies like DuPont in terms of expenditure on R&D inside the organisation and outsourced in strategic partnerships. That is the first question.
The second one is about another R&D strategy that some companies, and undoubtedly your own also, are pursuing. That is, to not necessarily form the relationship early on, say, with a university research capacity but to wait until it has spun out and then see whether it is worth acquiring it. Where do you think those two different strategies will mix and what will be the appropriate levels of them?
Tom Connelly: Two very interesting questions. First, with regard to the inside versus outside research, I would say the current situation in DuPont is that our external research budget is small, maybe not even 5 per cent. I would like to see it as multiples of that, 10 to 15 per cent. I showed on the chart that we are increasing our R&D footprint in developing markets - we need to do that as we build major technology centres in some of our new geographies. So I have an expanding internal footprint. We are shrinking selectively in some places our more developed markets, but the question for anybody with an administrative role here is how to free up money in budgets as our internal footprint is expanding, to dedicate more of our resources outside. That is one of the challenges we are living with today, but directionally we definitely need to do more outside, probably significantly more, than we are.
I would be happy with 10 per cent as an interim objective for how much we would spend outside. Some companies are talking about 20 per cent or a third or more, but for where we are and what we need to do, I'd be happy to double it to 10 per cent as at least an interim goal.
On the question as to at what point you access technology coming out of a university - whether to engage the university while it is in the laboratories or wait until they have partnered and set up a company and do it there - I don't think there are any hard and fast rules. There are some faculty members at universities that are not going to set up the company or would be quite happy at an early stage to have a company come in and pick up a piece of technology, take it literally out of their laboratories and move it toward commercialisation. There are others with a very strong entrepreneurial bent who are intent on setting up their own companies and trying to benefit from the multiple that is associated with taking a company and selling it to a strategic buyer at a later stage.
There are also organisations that try to help academics get to that point - venture capital groups that specialise in early-stage ideas, the kinds of things that come out of universities to really help a professor, maybe at a kind of mezzanine level, get the idea into a form where it is more attractive to a strategic buyer such as DuPont.
I don't think there are any hard and fast rules. It is really as much about the attitude of the people who have the technology as anything else.
Question 3: Following on from your previous comments about collaboration, that individuals in universities might be interested in perhaps starting their own company or venture capital or something, I wondered whether you might comment in general terms about ironing out some of the ideas used within the collaboration, particularly when I assume there might be researchers involved in the collaboration who themselves might have aspirations toward eventually starting a company or marketing their own research, while at the same time trying to balance this with the needs of the university and DuPont.
Tom Connelly: Again IP is often thrown up as a barrier to moving forward. It can be, but really it shouldn't be. But I made no distinction between working with an industrial partner and working with an academic institution. There are some companies that are difficult to work with on IP issues and there are some universities or some professors who are difficult to work with. My rule of thumb is to find people who are good partners and then you really don't have those kinds of issues.
I have come across things where we are sponsoring a piece of research and then they come back and tell you the invention was made on a Monday and that's the one day of the week that they don't work on the DuPont money! I may have one or two of those experiences and need to go elsewhere.
I have never viewed intellectual property as a barrier. It is something where you need to hammer out the agreements and make it very clear what belongs to whom, but we do the same thing with industrial partners.
Question 4: What about universities in India and China? Where are they on the learning curve of corporate collaboration?
Tom Connelly: It's an interesting question. I was telling some people earlier that I have been visiting universities in China for 20 years or more, and 20 years ago we were polite and they were polite but there was nothing going on of interest. It is shocking to me to see how quickly and dramatically that situation has changed. I visit leading research universities in China today and they tee up a dozen relevant subjects for my company, and they have faculty staff trained in the US or Australia - 80 per cent of their faculty have been trained outside China. They understand how to do research and they are doing leading-edge stuff.
We have had discussions in China but I would say that we have done more research with Indian institutes of technology and universities. We have a track record there, I would say at this point, that works well. We don't have that track record yet in China, but it is something that is going to come. Following the start-up of our own research facilities there, I have had interactions with the Minister of Science and Technology, the Chinese Academy of Sciences and leading research universities in the Shanghai area. There are real possibilities, but we don't have a track record yet.
Question 5: This is a left-field question, but if you were in charge of a country, what is the one thing you would do differently in funding the university sector, in promoting innovation?
Tom Connelly: I anticipated a lot of questions, but I just hadn't thought of that one. I am a visitor here and I don't really have a strong sense of the state of play.
To talk about my own research experience: funding is clearly a key issue, deciding how much you are going to spend, and then continuity of funding. I see this from my own situation in industry. My Chairman has never been lavish in the support of research but he has been consistent in his support of research. I don't mind. I like to operate in a resource-constrained environment; it really forces you to make choices and make sure that you are supporting the programs. But you need that continuity of funding to make sure that the things that you are really backing are able to go on.
The US issue is not spending enough on research, particularly in the natural sciences and engineering. The National Institutes of Health have seemingly got more than the natural sciences lately. So I think in the US the question is of spending more, especially spending more in the natural science areas. I don't know what the state of play is here in Australia.
Kurt Lambeck: The answer is again long-term consistent funds rather than short-term ones.
Question 6: I have two questions, but related. You credited the Bayh-Dole Act of 1980 with essentially underlying the creation of technology transfer, but interestingly most Australian researchers would see the Bayh-Dole Act as actually hindering collaboration between Australian researchers and those in the United States, for a number of reasons. This became a topic for particular attention during the negotiations over the Free Trade Agreement.
So, first of all, I would like to know whether you think our perceptions of that Act are wrong. Secondly, has the Free Trade Agreement had any impact at all on the operations of DuPont with Australia?
Tom Connelly: That second question is going to our President of DuPont Australia, so I will give him a moment to think about that one.
To go back to the Bayh-Dole Act, I think it had some beneficial effects, it had some unintended consequences. It was there to promote work between universities and industry, but it may have had detrimental effects on university to university collaboration, because if you thought you had something of value you wouldn't just talk to the people who wouldn't necessarily help you elaborate that technology but wanted a piece of it; you would want to talk to the people who wanted to buy the technology. So I can well imagine that there may be those kinds of effects.
When I met with the university presidents a while back, they were pushing for a study of Bayh-Dole plus 30 - what are the lessons learned from the Bayh-Dole Act, what has worked and what hasn't worked? I do think it did spawn a number of tech transfer offices in universities that were very, very aggressive in their approach to that. I am in the midst of one of those cases right now and it can be kind of unpleasant. But, as I said earlier, when confronted with those situations you just know that you're not going back. You just move on to find people who are more reasonable in their approach.
Hutch Ranck: I think we would all agree that we haven't seen the full positive effect yet on the Australian economy from the Free Trade Agreement.
Question 7: I wonder if I could put a spoke in the wheel, as it were, or kick the spoke a little bit. There are two questions that occur to me. One comes from talking to other large companies about collaborations with research institutions in general, and the deep concern that has been expressed to me by large companies, about the disclosure of their own strategy in developing collaborations with institutions. How much does that affect the way in which you disclose the kind of information you have been disclosing in the slides, for example, in regard to what your future strategies may be and therefore what your future interactions may be with other institutions unnamed?
The second question sort of relates to that, and this is where I get very negative, I suppose. You have portrayed a number of success stories, but research and development is full of failures, especially in the commercialisation context. I wonder if you would care to elaborate the level of failure that occurs, as a percentage relative to the number of successes that occur, in the sorts of collaborations you have carried out in the last 20 years.
Tom Connelly: I will answer the first question with a kind of vignette that I meant to mention when I was talking about the DuPont MIT Alliance. There was a kind of induction period of a year or a year and a half before we really started to get somewhere with that. And I would characterise that as the phase where the MIT researchers were repackaging the research they were doing and trying to gussy it up for funding through the DuPont MIT Alliance. At the same time, the DuPont scientists were describing some model system and saying, in effect: 'We want you to work on this, because we're not going to tell you what's really important to us. You just work over there, and maybe I will learn something over there that will help me with what I really want to do.' It took about a year to a year and a half for both of those bad behaviours to melt away. We don't have time to have somebody working on something and be paying for it if it is not important to us. It really took the relationships to develop before you could sit down and talk to people about things that were important.
I was sitting up at MIT next to one of the professors, listening to his student talk, and he leaned over and said: 'You know, if this one hits, this could pay for the whole investment that we have made.' And I'd had that thought, but what really struck me was that this professor had had that thought as well. You get to the point where you can get so caught up in not disclosing things that you never get anything done. I don't think my biggest risk is disclosing a little bit of my strategy to somebody who wants to work with me, so I'm not going to worry too much about that one for now. I think the greater risk is that we will play it so close to the vest that we get nothing out of the collaboration.
With regard to successes and failures, most things fail. Absolutely, most things fail. Our projects fail because we picked the wrong target, or we didn't get that market insight so we just went after the wrong thing and we failed. We fail because we didn't do the science right, we couldn't deliver technically that we wanted to. And also we can pick the right target and develop the right technology, but if we can't get it into the marketplace we can fail in the commercialisation phase.
And I think if you talk to any of our partners they will say their failures are onside with successes, and that is just the nature of research.
Kurt Lambeck: I think that is an appropriate time to stop the process for this evening! Thank you again very much, Tom.
Tom Connelly: Thank you very much. I appreciate it.
