agricultural engineering research. Much of the fundamental R&D that underpinned the gains in agricultural productiv­ity in NAE was publicly funded between the 1950s and early 1980s. Much of this stock of R&D knowledge and the re­search capability that provided it has now either been used up or depleted. Less is available for future needs.
     Government funded research is now largely confined to addressing non-market social, environmental and strategic issues as well as supporting fundamental research that would not attract private funding. Examples include research into livestock systems to reduce environmental burdens and re­search programs to assist farmers in disadvantaged, upland areas. Under the new paradigm of multifunctional, sustain­able agriculture, it is clear that there will be a continued need for government funding of R&D in the public interest.
     Spending on private sector agricultural research has grown relatively rapidly over the last 25 years and now ex­ceeds public spending in many developed countries (Alston, 2000). An additional and growing source of funding is that of nongovernmental, not-for-profit organizations that spon­sor research in pursuit of an organizational agenda, mostly associated with social, environmental, ethical or political objectives. They also provide a conduit for private or gov­ernment funds. Private benefactors also channel funds into trusts that pursue selected themes. In the future, increased emphasis on the multiple functions of agriculture is likely to call on a greater range of funding sources.
     The paradigm shift in agriculture towards multifunc-tionality and the concomitant shift in AKST have major implications for the provisioning of R&D in terms of prior­ity setting, funding and delivery mechanisms. Continuing reform of agricultural policy throughout the NAE region is likely to promote greater market orientation for agricul­ture, implying that governments will further retract from R&D that is "near market", leaving this largely to the pri­vate sector. Government funding of R&D is likely to focus on aspects of public good, addressing strategic issues such as food security, impacts of climate change, the long-term sustainability of agricultural systems and the protection of natural resources, the environment and the livelihoods of vulnerable rural communities.
     Where private R&D initiatives fail to respond to market potential because of high costs, high risks or long investment periods, governments can collaborate with private partners to underwrite commercial risks if they perceive a potential net gain in social and economic welfare. Collaborative fund­ing of R&D to support bioenergy cropping and processing systems is a case in point. Where technology change is pol­icy-induced, there is a strong case for collaborative public-private funding mechanisms for R&D. The EU Integrated Pollution Control Regulations and the Water Framework Directive are cases in point, justifying collaborative R&D ventures that share the burden of costs associated with new regulations.

6.5.3 Undertaking agricultural R&D
There has been a recent tendency to separate R&D funding and delivery mechanisms, with increased "contracting out" of government-funded research, diversifying the range of organizations engaged in research. As agricultural enhance-

ment has become less important as a policy goal, direct gov­ernment involvement in R&D to improve productivity has declined: it is now regarded as too "near market." As a re­sult, the number and size of government research institutes in some parts of NAE have decreased. In other cases, spe­cialist government research institutes and universities have increased their share of private and NGO funded research, utilizing specialist skills and facilities. Funding and deliv­ery regimes come together to provide a range of options for R&D management, considering the three main players: government, NGOs and private organizations (Table 6-4). Universities are a key delivery agent.
     There is now much greater diversity in the provisioning of agricultural research, especially regarding biosciences, with a growth in public-private funding partnerships among industry, NGOs and government, often involving universi­ties as research contractors. Potential complementarities include joint funding, pooling of facilities and expertise (including research management), economies of scale and learning, risk sharing and dissemination and commercializa­tion of research products into research outcomes.

6.5.4 Deciding on R&D
Regarding public goods, this is clearly a role for government and intergovernmental development agencies, engaging the key stakeholders in the process. Most NAE governments have R&D priority setting regimes in place and these will be increasingly linked to strategies to promote sustainable agriculture (e.g., DEFRA, 2002b, 2005b).
     Regarding private good aspects, decisions rest on the commercial considerations of business enterprises. Public-private partnerships can help to underwrite private R&D investment costs where risks are high and the development of successful research capabilities and products can signif­icantly enhance the public good. At a national level, this may include improving international competitive advantage (Gopinath et al., 1997; Ball et al., 2001).

6.5.5 Institutional arrangements and collaboration
The organizational arrangements  for identifying, priori­tizing, funding and carrying out R&D and, not least, the transposition of research outputs into knowledge, products and processes for adoption by end users, are critical to the overall successful outcomes of R&D. Again, this will reflect the dominant purposes to be achieved and, as far as serv­ing the public good is concerned, it is a responsibility of Government to provide an institutional framework within which various stakeholders can interact.
     R&D management includes arrangements for identifi­cation of needs, priority setting, pre-investment appraisal, research  procurement,  dissemination  and  follow-up.  As mentioned in earlier sections, it is imperative that key stake­holders are involved throughout this process to ensure R&D is relevant to end-user constituencies. The latter also im­plies full integration with the processes of "knowledge ex­change", including those of advisory and extension services. There are also important links with other services that affect technology adoption, notably credit and marketing.
     The further development of funding arrangements could be designed to promote enhanced cooperation not only