Innovation, Productivity and Sustainability in Food and Agriculture

Chapter 5. Innovation policy and the agricultural innovation system

Abstract

This chapter provides an overview of agricultural innovation systems in the reviewed countries. It outlines their diversity in terms of actors, ambitions, governance, funding mechanisms, incentives to invest and collaborate in research and innovation, and mechanisms to encourage adoption, including farm advisory systems. It also acknowledges efforts to improve the impact of public expenditure and make the system more collaborative and demand driven to increase relevance and thus adoption. Finally, it identifies main knowledge gaps in this area and consolidates recommendations to strengthen agricultural innovation systems made in country reviews.

• Agricultural innovation systems in reviewed countries differ in terms of ambitions, size and institutions. There is no one size fits all.

• One important trend in innovation policy is to improve the impact of public expenditure and make the system more collaborative and demand driven to increase relevance and thus adoption. However, top-down approaches continue to dominate in most countries.

• Efforts to improve the governance of the agricultural innovation system focus on developing more coherent and longer term strategies for innovation in food and agriculture, involving stakeholders more formally and from earlier stages, and strengthening evaluation frameworks.

  • Efforts to integrate farmers in the innovation process are growing in some countries.

  • Better information is needed to improve evaluation of research and facilitate priority setting.

• Agricultural research is changing but remains dominated by the public sector in most reviewed countries, while private research traditionally focuses on specific sectors like genetic improvements, fertilisers and chemicals, buildings and machineries, and food processing.

  • Various trends in public funding for agricultural research are found across countries: they decreased in the last 15 years in major exporting countries like Canada, the United States and the Netherlands; but increased in several other countries.

  • Research funding mechanisms also change, with the development of competitive project funding.

  • Government support research and innovation in firms through investment support and tax rebates, and by enforcing Intellectual Property Rights.

• Efforts to foster public-private collaboration have been made in most countries, but the capacity of local agri-food companies to engage in research activities remains often limited.

• International co-operation is important for countries of all sizes, to reduce costs and pool resources and exploit synergies on regional or global challenges. It is most developed among EU Member States.

• Renewed policy attention is paid on improving the adoption of innovation in farms and firms, through improvements in the enabling environment, and specific investment support.

• Farm advisory systems play an important role in innovation. They are in transition to adapt to new needs and provide a wider range of advice requiring re-training and flexibility. New intermediary actors emerge to meet these needs.

Diversity of actors and systems

Agricultural innovation systems involve a wide range of actors, including policy-makers, researchers, teachers, advisors, farmers, private companies, non-profit organisations, and consumers.

In most countries, governments provide strategic guidance for research and innovation, and provide funding to research institutions, private companies, and advisory systems. Government funding can be granted directly to research institutions, or through funding organisations, such as research councils or foundations, which allocate funds to projects. Depending on the country, part or all of agricultural research funding is integrated into the general innovation policy.

In several countries, the Ministry in charge of agriculture plays a prominent role in funding and performing agricultural research via agricultural research institutes that are part or under the umbrella of the ministry (e.g. INTA in Argentina, Embrapa in Brazil, Agriculture and Agri-Food Canada's research centres, Corpoica in Colombia, and the USDA Agricultural Research Service). These research institutes include regional offices and laboratories throughout the country, and are often active in knowledge diffusion.

Beside government research centres, universities in these countries also perform agricultural research and development (R&D), and have a significant role as in the case of land-grant universities the United States. They often focus their activities on regional agricultural strengths, and receive both federal and regional funds.

In the reviewed EU countries, agricultural R&D is funded as part of general R&D mechanisms, but performed mainly in universities with strong specialisation in agriculture or life sciences (University of Life Sciences in Estonia, Latvia University of Life Science and Technologies, Wageningen University in the Netherlands, and the University of Agricultural Science in Sweden). In Estonia, Latvia and the Netherlands, the agricultural university includes applied research institutes (e.g. Wageningen Economic Research). In Turkey, 43 universities are engaged in agricultural R&D, as well as two of the TÜBITAK institutes.1

Diversity is even higher among advisory systems, from competitive ones with a large range of suppliers for farmers to choose from, and minor government involvement, like in the Netherlands, to comprehensive government managed and funded systems like in Korea. In Brazil, the public advisory system focuses on non-commercial, smaller farmers.

Input industries play a large role in R&D for agriculture, in particular to improve variable inputs — seeds, fertilisers, pesticides, veterinary medicine — as well as farm machineries and equipment. However, these are relatively concentrated industries, which perform research in few, large countries. Some food processing companies are also active, but many lack capacity for research. Both upstream and downstream industries contribute to knowledge transfer. Increasingly, new companies emerge to provide specialised advice and services to farmers on new technologies. This is in particular the case with application of digital technologies to the sector (OECD, 2019b).

Governance of agricultural innovation systems

In all reviewed countries, strategic plans over five to seven years are developed for agriculture and food research, often as part of national science technology and innovation (STI) strategies and in co‑ordination between STI and agriculture-related ministries. In this framework, research organisations develop their own objectives. In countries with a diversity of public research organisations such as Canada, the People’s Republic of China (hereafter “China”), Turkey and the United States, co‑ordination is all the more important. An important recommendation in many reviews (e.g. the Netherlands and Sweden) was to define long-term objectives for R&D and innovation. The Netherlands has translated this into the development of a strategic knowledge and innovation agenda. Social issues are central to this agenda and will be implemented through multi-year mission-driven innovation programmes. Together, the government and stakeholders are defining concrete goals that will be pursued with a wide range of policy instruments.

Productivity growth remains an important objective of agricultural innovation systems in many countries, but the range of objectives has generally broadened to include sustainability and climate change issues, food and health and other societal issues. The new emphasis on more complex issues requires multidisciplinary and multi-sectoral approaches, thus the need to reinforce co-operation between researchers and research organisations. A particular challenge is to adopt a longer term perspective for long term challenges, such as climate change.

In defining objectives and allocating funding, it is particularly important to define clearly the respective roles of the public and private sectors, and areas of mutual interest and possible co-operation. An analysis of public and private R&D activities in the United States suggests there is complementarity, with public research focusing on different specific topics than private, and that private often building on the scientific findings of public research. It is less clear in countries with lower private investment in research, often linked to a market size too small to attract multinational companies or low capacity of national companies.

In most countries, approaches to agricultural innovation remain mainly top-down although efforts are increasing to better understand innovation needs and to strengthen participation of stakeholders, including in the definition of strategies and objectives. In a few countries, formal institutions have been created to facilitate discussion and co-operation throughout the innovation process. This is in particular the case in Canada with the Value Chain Round Tables (Box 5.1), in the Netherlands with the creation of Top Consortia as part of the implementation of the Top Sector Policy (Box 5.4), and in Australia with the Research and Development Corporations (RDCs) (Box 5.5). A further challenge is to ensure a wider representation of stakeholders, which would include the diversity of sectoral and societal interests.

In many countries, innovation is taking place within specific value chains. Argentina and Colombia provide examples of very successful value-chains investing in research and innovation, co-existing with less efficient ones.

Box 5.1. Value Chain Round Tables in Canada

Twelve Value Chain Round Tables (VCRTs) have been launched in 2003 to facilitate co-operation across the supply chains at the national level. Bringing together key industry leaders from across the value chain – input suppliers, producers, processors, food service industries, retailers, traders and associations (geographical regions and sector diversity are also considered) — with federal and provincial government policy makers, VCRTs have become central vehicles for: identifying sector strengths and weakness; capitalising on domestic and international market opportunities; sharing information and building trust across commodity sectors; identifying research, policy, regulatory and technical requirements; creating shared visions and co-operative long-term strategies; and responding to crises.

Source: OECD (2015a), Innovation, Agricultural Productivity and Sustainability in Canada, https://dx.doi.org/10.1787/9789264238541-en.

Evaluation of research and innovation policy is important to increase the efficiency with which public funds are used, and more broadly to improve the functioning of the research and innovation system and its contribution to addressing a wide range of socio-economic and environmental issues and to global challenges (Joly et al., 2016). Practices are very diverse among reviewed countries, and evaluation is often partial, focusing on financial considerations (Table 5.1). Research excellence is the main criteria for the evaluation of researchers, projects and institutions. As a result, networking and knowledge transfer activities, which increase impact, are not valued. In many countries, innovation policy or research organisations are at minimum assessed with regard to set objectives. Latvia requires for example international assessment of institutions. In the Netherlands, research institutions and top sectors evaluate their activities annually. In addition, an independent review is required every five years, based on quality of research.

Evaluation of research projects and researchers is also common practice. In the evaluation of researchers or research units, research excellence validated by scientific peers is the main objective. While this criterion is recognised at the international level, other types of incentives are needed to ensure better integration in the agricultural innovation system. Project evaluation responds to criteria set for project selection, and increasingly includes ex ante and ex post impact assessment.

Evaluation is often left to research organisations. Embrapa in Brazil estimates returns on investment. In Australia, the Commonwealth Scientific and Industrial Research Organisation (CSIRO) used independent assessment (e.g. ACIL-Allen Consulting, 2014), including estimation of returns on investment, to develop a new strategy and develop a common framework for all impact assessments within CSIRO. These guidelines notably develop a programme logic based on input-outcome-impact model. It is suggested to broaden the dimensions of impacts considered, including economy, environment and society, and to use a mix of methods for investigating the impact, including qualitative, cost-benefit analysis and option values (to account for the externalities) (CSIRO, 2015). Evaluation in the USDA Agricultural Research Service (ARS) is based on performance targets established in the 17 National Programmes. Most targets are on providing research outputs, but a few targets concern research outcomes or economic and environmental impacts. Both standard economic approaches, which aim to estimate economic benefits of research investments, and case-study approaches, which aim to analyse the processes of impact generation, are applied.

Table 5.1. Type of evaluation of innovation policy

	Basis	Level	Agricultural innovation system
Argentina	Project implementation	Research organisation	No
Australia	Objectives and impact, Returns on investment, environmental indicators being built, larger evaluation every 5 years	Organisations and policy	In development
Brazil	Returns on investment	Research organisation	No
Canada	Programme objectives	Programmes	No
China	..		No
Colombia	Achievements relative to objectives		No
Estonia	Targets of Innovation and agricultural policy	Policy, projects	No
Japan	Objectives in strategic plans and programmes	Policy, projects	No
Korea	Institutional objectives, and research quality	Policy, projects	No
Latvia	Targets of Innovation and agricultural policy international assessment of institutions	Policy, institution, projects	No
Netherlands	Objectives and quality of research, independent review every 5 years	Institutions and top sectors,	No
Sweden	Quality of research	By research institutions	No
Turkey	Planned objectives, quality of research	Projects	No
United States	Objectives and impact assessment, including environmental targets	ARS, projects	No

Note: ..: Not available.

Source: Country reviews; Joly et al. (2016), “Agricultural research impact assessment: Issues, methods and challenges”, https://dx.doi.org/10.1787/5339e165-en.

Public funding of agricultural R&D

Public investment on agricultural R&D as a percentage of agricultural value-added (or research intensity) ranges from less than 0.2% in Turkey to over 2.5% in Korea and Switzerland, when measured by Government budget allocation for R&D (GBARD), which includes public funding for R&D on agricultural sciences performed in both public and private organisations (Figure 5.1).2 Public research intensity is usually higher in agriculture than the overall economy (OECD, 2013).

Following significant increases over the period 2000-16, public research intensity is the highest in countries that provide high support levels to their farmers. It is also relatively high in Brazil and Canada which are large competitive exporters, but where private companies often lack capacity to invest in agricultural R&D. Conversely, Australia, the Netherlands and the United States rely to a greater extent on private funding of R&D and research partnerships (see below).

China's government expenditure on agricultural research has increased dramatically since 2000 and overtook US public expenditure in 2009 (in constant PPP USD) (Figure 7.6 in OECD, 2016). As a result, public research intensity has almost doubled in China since 2000, but remains below 1%.

Government expenditure on agriculture research has decreased over time in several major exporting countries (Figure 5.1). This could threaten the capacity of public research to cover areas of less interest to producers (longer-term, public goods) and engage in collaborative activities (e.g. at an international level). In countries with significant private R&D investment, this has not necessarily replaced public investment, as private and public R&D activities are expected to be complementary (OECD, 2016).

Figure 5.1. Public R&D intensity on agricultural sciences, 2000* and 2017*

Notes: * or nearest available year: 2017 is replaced by 2016 for Estonia, Japan, Korea, Latvia, and the United States; by 2015 for Switzerland; by 2014 for Canada; by 2013 for Brazil, China and Colombia. 2000 is replaced by 2002 for Estonia. 2000 data are not available for Turkey and Latvia.

1. Government budget allocations for R&D (GBARD) is a funder-based approach for reporting R&D, which involves identifying all the budget items that may support R&D activities and measuring or estimating their R&D content. It enables linking these budget lines to policy considerations through classification by socioeconomic objectives. However, it provides only a partial indicator of investment in public agricultural research, since it refers to research funding instruments dedicated specifically to agriculture.

Source: OECD (2019a), “Research and Development”, OECD Statistics (database), https://stats.oecd.org/ (accessed January 2019); and for Brazil, China and Colombia: ASTI (2017), Agricultural Science and Technology Indicators (database), www.asti.cgiar.org/data (accessed March 2018).

Statlink

Public R&D funding mechanisms are changing in an effort to improve relevance of agricultural R&D for the sector and with government objectives reflecting wider societal goals. Governments (and donors) increasingly use project-based, competitive, funding as an efficient way to allocate resources to priority areas. In most countries, governments use both block funding — i.e. research grants to research organisations, often based on past performance, and project-based funding, but to a different extent. In countries with a high share of project-based, competitive funding of R&D, researchers and research institutions experience significant transaction costs and instability.

In Brazil, the main agricultural research organisation under the ministry of agriculture, Embrapa, receives a large share of funds via block funding. At the opposite extreme, most public funds are allocated to competitive projects within top sectors in the Netherlands. Similarly, project-based funding represented over 90% in Estonia over the period 2005‑13. However, this raised concerns for long-term strategic planning and sustainability of R&D institutes, and the Estonian Government decided to gradually increase block funding to reach a 50-50 ratio. As a result, the share of block funding reached 20% of the total in 2015 and 27% in 2016. In Sweden the share of block funding has also increased to 45% of university funding in Sweden. Researchers in EU Member States participate in EU funded projects, which in some of them account for a significant share of R&D funding.

In the United States, it is difficult to evaluate the importance of block funding versus project-based funding, but the Agricultural Research Services of the US Department of Agriculture, which receives more than half of Federal expenditure on agricultural R&D, receives mainly block funding. In other reviewed countries, while precise information is not available, but Table 5.4, which attempts to summarise main policy parameters proposes likely ranges based on research organisations and funding institutions.

Private role in food and agricultural R&D

Private investment in food and agriculture R&D is difficult to track in many countries, as it is often missing or incomplete in official statistics. However, estimates of private spending on food and agricultural R&D suggests fast growth in private R&D worldwide, partly in response to higher world prices for some commodities after 2002 (Fuglie et al., 2011; Fuglie, 2016).3 In particular, there has been a spectacular increase in private investment in agricultural R&D in the United States in recent years. As a result, the share of public funding in the total has declined from around 50% in the early 2000s to 26% in 2014 (Figure 1.20 in OECD, 2018a, OECD, 2016). Private spending on agricultural R&D is mainly located in high-income countries (88% in 2014, compared with 94% in 1990) and highly concentrated in the largest firms.4

In the Netherlands, the share of private contributions to research projects under the two agriculture-related Top Sectors is about 30% (OECD, 2015b). This is equal to the estimated average share for agriculture R&D at the global level.5

Business research intensity for agriculture — as measured by Business Expenditures on R&D (BERD) as a percentage of gross value added — is the highest in the Netherlands, followed by Australia and Canada (Figure 5.2). Although data for the United States are not available in the OECD database, other sources indicate a high business research intensity for agriculture.

Among reviewed countries, research intensity in the food manufacturing industry is the highest in Japan, Korea, the Netherlands, and the United States (Figure 5.2). Day-Rubenstein and Fuglie (2011) find that highest rates are associated with the presence of large multinational companies. In many reviewed countries, national food processing companies often lack size and capacity to do research. This may be due to the cost of doing research locally and the small size of the market, which makes it difficult to develop new markets and products, but also to regulatory burdens and inconsistencies, and particular features of intellectual property protection as applied to rural R&D.

Figure 5.2. Research intensity in the agriculture and food and beverage processing industry, 2016*

Notes: * Or most recent available year; Food and beverage data are not available for Australia; Agriculture data are not available for Sweden, Switzerland and the United States.

1. Business Expenditure on R&D (BERD) is the measure of intramural R&D expenditures within the business enterprise sector (regardless the sources of R&D funds).

Source: OECD (2018b), National Accounts (database), [Value added and its components by activity, ISIC rev4], Research and Development Statistics (database), [Business enterprise R-D expenditure by industry (ISIC rev4)]; MSTI Main Science and Technology Indicators (database), [BERD as a percentage of GDP], https://stats.oecd.org/.

Statlink

Policy incentives to private investment in R&D and innovation

In most reviewed countries, governments encourage innovation activities in the private sector, including by fostering knowledge markets through protecting intellectual property rights (IPRs), providing direct or indirect financial incentives, engaging in public-private partnerships (PPP), and providing information and sharing the outcomes of public research (spill-overs).

Intellectual Property Rights

IPRs provide an important incentive to invest in innovation by enabling firms to recover their investment. Rights-holders can exclude competitors from use of an innovation for a limited period of time or, in the case of open innovation approaches, they can promote access and sharing. The challenge for IPR regulations is to provide incentives for private investment in innovation, without compromising the sharing of knowledge and further innovation (OECD, 2013).

In most reviewed countries, there are two main types of IPRs used in agriculture: patents and plant variety protection in the form of a breeder’s right. Duration of patent protection is generally 20 years. To facilitate the innovation process, Australia grants innovation patents with a shorter protection of eight years. The International Union for the Protection of New Varieties of Plants (UPOV) offers protection to the “breeder” of a plant variety, in the form of a “breeder's right”. This gives the developer of a new variety the right to exclude others from commercialisation, but allows farmers to use the seeds produced from a protected variety for subsequent plantings, and researchers to use them for further breeding research. The breeder's right under UPOV is granted for a period of not less than 20 years or, in the case of trees and vines, for not less than 25 years. In Estonia and Sweden, the duration of the right is longer (25 years and 30 years for trees). Argentina, Brazil, China, and Colombia signed the 1978 UPOV convention in the 1990s, but have not signed the 1991 one, which offers strengthened protection and improves the plant breeders’ ability to recover their development costs and to generate funds for re-investment. Most other countries have signed the 1991 UPOV convention around 2000, but Turkey and Switzerland signed in 2007-08 and Canada in 2015 after the publication of the OECD review.

Some reviews also mention the importance of trade secrets, which give indefinite protection, in the development of hybrid seeds (OECD, 2016) or the protection of food and drinks. Several reviews also mention geographical indications, which also provide perpetual protection, but without commenting on their impact on innovation.

According to the World Economic Forum (WEF) Global Competitiveness Indicator, intellectual property protection is high in most of the reviewed OECD countries, which are knowledge-based economies (Figure 5.3). Most of the increase took place in the 1980s and 1990s, Plant Variety Protection has also increased significantly around 2000 with the signature of the 1991 UPOV convention. In the non-OECD member countries that were reviewed, patent protection is lower, but has increased significantly in the last decade.

The strengthening of IPR protection in recent decades has been associated with an increase in private sector investment in agriculture-related research and development and a surge in innovation leading to improved plant varieties, agricultural chemicals, and production technologies. It has also provided farmers access to foreign innovations. Conversely, in Canada, delaying the adoption of UPOV91 (to 2015) compared to trading partners has limited Canadian farmers’ access to new, more productive varieties: it limited levels of investment in the domestic plant breeding programmes of some crops, and many foreign breeders did not seek Plant Breeder’s Right protection nor introduced their varieties into Canada.

Many reviews outline the significant impact of genetic improvement in agricultural productivity growth and the reinforcement of IPR protection in this development. Patent positions in combination with technological developments have led to large consolidation among breeding companies. The access barrier for new companies to the plant breeding sector is high, where IPR plays a role next to the large amount of knowledge and expertise required to set up a breeding company and the long development period for new varieties. New gene editing technics, however, become more accessible to smaller companies and shorten the process for developing new varieties. But in the absence of some breeding exemption under patent law entailing that anyone may make use of patented biological material for the purpose of breeding, or discovering and developing of other plant varieties without the consent of the patent holder, large breeding companies holding existing IPRs are likely to remain ahead (OECD, 2015b).

As noted in the Dutch review, however, not all innovations are or can be protected by IPRs, in particular non-technological innovations or in areas where fast adoption is required to maintain a competitive hedge, like horticulture (OECD, 2015b).

Most reviewed countries are members of the OECD Schemes for the Varietal Certification of Seed Moving in International Trade, which promotes the use of agriculture seed of consistently high quality, for most crops. Exceptions are China, Colombia and Korea. The OECD certification provides for official recognition of “quality-guaranteed” seed, thus facilitating international trade and contributing to the removal of technical trade barriers.6

Figure 5.3. Global Competitiveness Index: Intellectual property protection, 2007-08 and 2017-18

Note: Indices for EU28 and OECD are the simple average of member-country indices.

Source: World Economic Forum (2017), The Global Competitiveness Report 2017-2018: Full data Edition, http://reports.weforum.org/global-competitiveness-index-2017-2018/.

Statlink

Support to R&D investment

Mechanisms supporting private R&D investment are generally economy-wide and not specific to agriculture or food related activities. There is little information in country reviews on the extent to which this support benefits agri-food companies.7

Many countries increasingly support R&D through tax instruments (Figure 5.4). They include tax rebates on eligible income, profit, investment, exploitation costs, including cost of labour employed in R&D, or benefits from research projects, patents and royalties (e.g. Innovation Box in the Netherlands). The share of tax incentives in R&D support has been increasing in recent years. In some countries like Canada, most support to R&D is through tax incentives, while it is hardly or not at all used in others such as Estonia and Sweden.

The extent to which tax subsidies encourage additional R&D activities and reach intended beneficiaries, e.g. small and medium-sized enterprises (SMEs), which would not carry out R&D otherwise, needs to be investigated. They tend to favour companies, which make profit or have the capacity to carry out R&D activities, while the food and agriculture sector includes a high number of smaller companies with little capacity for research. Direct support gives more scope for targeting them.

Most countries provide direct support to R&D investment in private companies, including through project funding and procurement mechanisms. Few mechanisms supporting innovation in private companies are sector-specific. In many countries, some target innovation in SMEs (e.g. the Small Business Innovation Research (SBIR) programme in the Netherlands and the United States), but the extent to which this support benefits the food industry is unclear. General mechanisms may include specific support earmarked for food and agriculture-related innovation activities, sometimes as part of a strategy (e.g. bioeconomy, genomics).

In some countries, agricultural policy includes support for participation in innovation partnerships or networks, as in the EU Rural Development Programme. In Canada, two programmes within AgriInnovation are geared towards encouraging industry-led innovation and investment: Agri-Science Cluster and Agri-Science Projects (Box 5.1).8

Demand-driven mechanisms are an innovative and promising way to fund research. Reviews mention the use of procurement or pull-mechanisms9 for innovation in general, and some more specifically in food and agriculture. In the Netherlands for example, among the diverse business innovation funding instruments, two deliver project funding for public procurement addressing societal challenges (Table 7.4 in OECD, 2015b). Latvia developed regulations to facilitate green procurement, which should reduce the environmental impact of procured goods, and foster the development of environmentally-friendly goods and services. However, experience in the food and agricultural sector remains limited.

Figure 5.4. Change in government support for business R&D through direct funding and tax incentives, 2006 and 2015

Note: There are no tax incentives for R&D in Estonia and Sweden, while in Latvia, they were introduced in 2014.

Source: OECD (2017), R&D Tax Incentive Indicators, https://oe.cd/rdtax (accessed July 2017).

Statlink

Public-private co-operation for innovation

Co-operation between various public and private actors in the agricultural innovation system is essential to increase the returns from public funds and to tailor innovations to needs. Public-Private Partnerships (PPPs) are one policy option, which may facilitate improved outcomes but governance, design and implementation issues need to be carefully considered to ensure success (Box 5.2).

Reviewed countries use different institutional and funding mechanisms — public funding to research projects requiring public and private participation and co-funding, foundations, institutions. Table 5.2 provides examples of mechanisms to foster PPPs in reviewed countries. Most of them are not sector-specific, but apply to food and agriculture R&D. There are few examples of agricultural-specific mechanisms.

Box 5.2. Governance and implementation considerations for successful Public-Private Partnerships (PPPs)

• Setting clear objectives and rules, and implementing regular monitoring and evaluation is essential.

• Evaluation procedures need to be well-tested and include impact analysis.

• Transparency, consultation with stakeholders and the establishment of dispute settlement and exit strategies are also important.

• Institutional arrangements need to be clear, including the sharing of costs and benefits, particularly the terms governing the sharing of Intellectual Property Rights (IPRs) between partners.

• Once government has determined its priority areas, PPPs should be selected using a transparent, open and competitive process. Value for money is the main criteria but common aims, mutual benefits and complementarity among the partners are also important.

• Improving the capacity of partners is an important factor of success, and is particularly relevant for agricultural innovation. Providing specific education, training and advice to improve skills for PPPs management would help in this regard.

Source: Moreddu (2016), “Public-Private Partnerships for Agricultural Innovation: Lessons From Recent Experiences”, https://doi.org/10.1787/5jm55j9p9rmx-en.

There is evidence that the number of PPP agreements is increasing in Brazil and the United States. The Agricultural Research Service (ARS) of the US Department of Agriculture (USDA) is engaged in R&D partnerships to address major challenges such as climate change, bioenergy, food security, pests, and water use. Moreover, an agriculture-specific institution, the Foundation for Food and Agricultural Research (FFAR), was created in 2014, as an independent, board-driven, non-profit organisation, to foster collaboration between government, university, industry, and non-profit researchers (Toole, 2014). PPPs are based on Cooperative Research and Development Agreements (CRADA). A CRADA is a written agreement between a private company and a government agency to work together on a R&D project, which allows both parties to keep research results confidential for up to five years under the Freedom of Information Act. It allows the government and the partner to share patents and patent licenses and permits one partner to retain exclusive rights to a patent or patent license. In 2016, the Ministry of agriculture, forestry and fisheries of Japan launched the Council of Industry-Academia-Government Collaboration for “The Field for Knowledge Integration and Innovation (FKII)”, which aims to be a cross-sectoral platform of people, information and funds for agricultural research (Box 5.3).

Table 5.2. Public-Private Partnership programmes for R&D, selected countries

	Name	Duration	Responsibility	Description
Australia	Co-operative research Centres (CRCs)	Since 1990	Industry	Partnerships between different research funders, suppliers and end users formed to undertake R&D in specific areas, with a particular emphasis on applied R&D
	Research and Development Corporation (RDC)		Rural	Industry (farmers)-public co-finance and have a say in what research is done but farmers do not conduct research
Brazil	Legal Framework and funding mechanisms	2004	Research	Facilitate the participation of public entities in co‑operation activities, and sharing of equipment
Canada	Network of Centres of Excellence (NCE)	Since 1989	Science-research	NCEs help mobilise multi-disciplinary research capacity, create large-scale, academic-led research networks, and engage public and private partners
	AgriInnovation	Since 2013	Food and agriculture	As part of the Industry-led Research and Development stream, agri-Science Cluster support aims to mobilise and co-ordinate a critical mass of scientific expertise in industry, academia and government (Box 4.1)
Estonia	Competence Centre (CC) programme	Since 2004	Enterprise Estonia	Competence centre: private entities established by a consortium of R&D institutions and enterprises (Box 7.3 in OECD, 2018c). Three out of six CC concern the food and agricultural sector1
	Innovation and development vouchers	Since 2008	Enterprise Estonia	Provides SMEs with grants for co-operating with higher education institutes, test laboratories or Intellectual Property experts to develop innovative solutions
	Cluster programme	Since 2008	Enterprise Estonia	No agricultural enterprises were eligible
European Union	RDP measure co-operation	Since 2014	Agriculture	EU Member States can choose to provide funds to R&D co-operation projects
Japan	Field for Knowledge Integration and Innovation	Since 2016	Agriculture	A cross-sectoral platform of people, information and funds in agricultural research (Box 5.3)
Netherlands	Top-Sector policy	Since 2011		Box 5.4
Sweden	VINN excellence Centre	2003-18	Industry-Energy-Innovation	Competence centre
United States	Engineering research centres	Since 1985	Science	Competence Centre
	Industry-University Co-operative Research Centre	Since 1979	Science	Competence Centre
	Small Business Innovation Research (SBIR) programme		Small Business Administration	Funding programme for small business to engage in Federal R&D – with potential for commercialisation
	Small Business Technology Transfer (STTR)		Small Business Administration	Funding programme to facilitate co-operative R&D between small business concerns and US research institutions – with potential for commercialisation
	Foundation for Food and Agricultural Research (FFAR)		Agriculture	Independent, board-driven, non-profit organisation created to foster collaboration between government, university, industry, and non-profit researchers. Funding is distributable only with an equal amount of non-Federal matching funds

1. There is also a regional CC and a Centre of Excellence with activities related to plants.

Source: Moreddu (2016), “Public-Private Partnerships for Agricultural Innovation: Lessons from Recent Experiences”, https://dx.doi.org/10.1787/5jm55j9p9rmx-en.

Box 5.3. Platform for open innovation in agriculture in Japan

In 2016, The Ministry of agriculture, forestry and fisheries of Japan launched the Council of Industry-Academia-Government Collaboration for “The Field for Knowledge Integration and Innovation (FKII)”. FKII aims to be a cross-sectoral platform of people, information and funds for agricultural research. The FKII is composed of three-layers:

• The Council of Industry-Academia-Government Collaboration that exchanges information among members such as producers, private companies, universities, research organizations.

• A R&D Platform that engages in collaborative research led by a designated research director.

• A Research Consortium that perform joint research.

As of May 2018, FKII includes 1 751 organisations and 690 individual members in the council, and launched 118 R&D platforms.

Cross-sectoral collaboration through FKII is expected to promote the commercialisation of new technology and enhance incentives for private investment in agricultural R&D. Most of the private companies engaged in R&D related only to the merchandising or commercialisation of their own products or services. Universities and public research organisations have been playing a central role in basic and applied research, which requires a long time horizon before results are commercialised.

FKII is intended to connect each R&D stages with the participation of diverse actors of the Agricultural Innovation System. The financial support through FKII focuses on the commercialisation of research output from basic to applied, in a three-to five-year period.

NARO Bio-oriented Technology Research Advancement Institution (BRAIN) provides support for R&D that adopts a new support mechanism via a matching fund method designed to encourage collaboration with private-sector companies, as a model R&D project that uses new open innovations conducted in the FKII. A total of 17 agricultural, forestry and fisheries projects were adopted in the model projects that were offered from public proposals in 2016/17. They include, for example, “development of a model plant factory system for the Asian monsoon region” and “development of an AI robot-operated greenhouse.”

Source: MAFF (2016), Field for Knowledge Integration and Innovation: Organization and Evolution Since Fiscal Year 2016, www.knowledge.maff.go.jp/uploads/0f46f5c7b37748f264f227c4073ffd134b453ba3.pdf.

The Netherlands went the furthest in placing PPPs at the heart of the R&D strategy introduced in 2011 — the Top Sector policy (Box 5.4). Nine key sectors have been identified with strong market positions. Two Top Sectors are dedicated respectively to the export orientated agri-food sector, and the horticulture and propagation materials sector. The aims are to maintain the competitive edge of these sectors, through innovation; increase private investment in pre-competitive research and foster networking. However, by giving the industry a leading role in setting innovation agendas, this policy risks focusing public funds on low risk and short-term R&D activities, away from research with more fundamental, public goods aspects needed to address long-term challenges. As originally implemented, the policy raised concerns over the long-term capacity of the agricultural innovation system. It also appears that while cost sharing is 50-50 in the agriculture-related top sectors, the effective private contribution is smaller when investment support and tax rebates are taken into account (OECD, 2015b).

Box 5.4. The Top Sector policy in the Netherlands

Implementation

The Top Sector policy subjects the granting of public funding to participation in PPPs within top sectors and gives industry a leading role in setting innovation agendas. Public funds have to be matched with an equivalent contribution from the private sector (50-50), which can be in kind (access to facilities) or financial, in which case it can benefit from public support (investment or tax rebates).

In the Top Sector policy the business community sets together with the government and scientists the agenda for R&D investments in its field. The government invites businesses and scientists to draw up action plans, which serve as a base to develop concrete lines of actions.

Each top sector has created one or more Top Consortia (TKI) for knowledge and innovation, where entrepreneurs and researchers work together in innovative products and concepts. Every TKI has a board with members from business and knowledge institutes. The government is an observer in the board. Programming is done by calls to tender leading to a number of identifiable projects since 2012. The activities of top sectors are regularly monitored through TKI's reports.

The basis for the implementation of innovation activities is the Innovation Contract. Each Top Sector draws up an Innovation Contract, in which researchers, entrepreneurs and the governments (represented in the so-called Top Team) agree on measures (mix of fundamental research, applied research, valorisation), plans to develop innovative products and services, and financial contributions.

Initial findings

One original objective of the Top Sector Policy was to leverage business-sector R&D and increase the applicability of public research. While companies participating in top sectors already invested in innovation, public co-funding focussing on pre-competitive research was expected to reinforce their contribution in this area. Early findings suggest companies, including multinational ones, increased investment in pre-competitive research, but that total private expenditures did not increase overall.

The policy was also expected to promote closer co-operation between knowledge institutes, public authorities and business. In the food industry, they improved co-operation between the processing and retail levels, as co-operation already existed among other components of the chain. All top sectors have a human capital agenda meant to strengthen the linkages between education institutes (from vocational training to university) in order to meet the needs of the sector itself.

The PPP approach is also to facilitate the marketing and adoption of innovation, and reduce the technological gap between small and large companies through knowledge transfer, as quality systems become more complex.

Source: OECD (2015b), Innovation, Agricultural Productivity and Sustainability in the Netherlands, https://dx.doi.org/10.1787/9789264238473-en.

Overall, support to private R&D and PPPs often benefits multinational companies to a greater extent than smaller enterprises as they have a greater research capacity. Some programmes, however, target specifically smaller companies.

In a few countries, farmers contribute to the funding of agricultural R&D via statutory or voluntary levies. Some processors also pay levies (red meat processors, timber mill operators, sugar mills, and winemakers). Levy funding ensures that research is adapted to industry needs and thus widely adopted, but favours major, well-organised commodity sectors. The Australian Research and Development Corporation (RDC) model, based on 50-50 co-funding by farmers and the government, channels a large part of agricultural R&D funding (Box 5.5). In some RDCs, the industry levies exceed the matching cap. While organised by commodity sectors, some RDCs have wide coverage, including small and emerging industries (OECD, 2015c). In Canada, mandatory levies (check-off) are used to financially support both marketing and research activities for a variety of farm products. A majority of check-off systems is implemented and managed at the provincial level. In Colombia, research activities conducted by producer associations are funded through commodity taxes levied on private sector production or exports. Thirteen producer associations are involved in agricultural research. Some producer associations have their own research facilities, called “supply chain research centres” (CENIs), and conduct their own research (OECD, 2015d). The four main research centres are for coffee, palm oil, sugar cane and rice. In Sweden, the Federation of Swedish Farmers (LRF) created in 1996 the Swedish Farmers’ Foundation for Agricultural Research as an independent legal organisation receiving funding from both the LRF and the government. About SEK 57 million are distributed every year in support of agricultural needs-driven research, of which about two-thirds from private sources. In total the Foundation budget is equivalent to around 13% of total government funding on agricultural R&D (OECD, 2018d).

Box 5.5. The Australian rural Research and Development Corporation (RDC) model

A unique feature of the Australian rural innovation system is the Rural Research and Development Corporation (RDC) model of co-financing of rural R&D activities established in 1989. It places interactions between public R&D and agricultural industries at the heart of the rural innovation system and has channelled in recent years a significant share of Australian Government spending on rural research and development (R&D).

Under the RDC model, the Australian Government matches industry R&D funds collected from primary producers via statutory or voluntary levies, dollar for dollar, with maximum matching contribution per year of 0.5% of an industry’s gross value of production.

This co-investment model:

• Generates greater spending capacity.

• Ensures that producers who benefit from research contribute to its costs.

• Ensures that research is of practical value.

• Facilitates greater and faster uptake of research outputs.

R&D activities within RDCs involve various rural and general R&D institutions and are organised by commodity, although some RDCs address broader challenges at supply-chain level. Originally competitive- and market-driven, the model has become more collaborative and inclusive. While the RDC model does not directly integrate agribusiness processing and retailing stakeholders in funding decisions, potentially limiting capacity to respond to demand for product and process development along the food chain, many RDCs undertake extensive marketing aspects including trade access and working along the supply chain to determine product and process development needs. By design, the RDC model is more adapted to marginal improvements than fundamental changes in production systems and resource management. Past evaluations have questioned the complex arrangements and unclear funding flows making evaluation difficult. They have also found that the overall level of public support for industry-focused research too high, outlining that the basis for the Government’s matching contribution to RDCs provides no incentive for producers to increase their investments in the model over time (Productivity Commission, 2011).

Source: OECD (2015c), Innovation, Agricultural Productivity and Sustainability in Australia, https://dx.doi.org/10.1787/9789264238367-en, based on Productivity Commission (2011), Rural Research and Development Corporations, Report No. 52, www.pc.gov.au/projects/inquiry/rural-research.

Farm advisory systems

Farm advisory systems have had an important role in the transfer and successful adoption of innovation, in particular at early stages of development. There is a wide diversity of systems, public and private providers and funding mechanisms within and across reviewed countries (Table 5.3).

The role of the government varies from being the main funder and provider like in Japan and Korea, to co-funding and guiding services managed by independent organisations like in Estonia. In some countries, farmers' organisations play an important role in providing advice to farmers, who pay collectively or individually for services. In the Netherlands, the national advisory system was privatised and replaced by a diversity of private providers. Consulting firms emerged also in different countries, in particular for specialised knowledge such as management or Information and Communication Technologies (ICTs). Small amounts of subsidies are available in the Rural Development Plan (RDP) for farmers to access services. In Brazil, public provision is only for smaller farms, while large commercial farms are expected to pay for the service.

Table 5.3. Examples of advisory services

	Main institutions	Source of funds	Countries
State-run	Public organisations at regional and national level	Wholly financed from public funds	Brazil for smaller farms, Colombia, Japan, Korea, Sweden, Turkey, United States
Public-Private Service	Increasingly provided by private consultant firms	Farmers partly or wholly pay for services; centralised and decentralised	Canada, China, Estonia, Australia, United States
Farmers’ Organisations	Farmers’ organisations	Membership fees and payments by farmers	Australia, Canada, Colombia, Japan, United States
Commercial	Commercial firms or private individuals	Payment through project implementation or grants	Netherlands, commercial farms in Brazil, Turkey, United States

Note: Several systems co-exist in some countries.

Source: Adapted from OECD (2013), Agricultural Innovation Systems: A Framework for Analysing the Role of the Government, https://dx.doi.org/10.1787/9789264200593-en; and OECD (2015e), Fostering Green Growth in Agriculture: The Role of Training, Advisory Services and Extension Initiatives, https://dx.doi.org/10.1787/9789264232198-en.

In addition to established advisory services, farmers receive advice from input suppliers, downstream industries (in particular with integrators, labels, organic) and co‑operatives.

Farm advisory services are in transition to meet new demands without increasing costs. A few trends can be observed:

• Farm advisory services need to cover a broader range of subjects and respond to wider and more complex issues. Beside technological innovation to increase productivity growth and competitiveness, farmers increasingly need advice to implement more sustainable practices, better manage their farm, and market their products. Advice needed is thus less commodity specific, more focused on production systems and environmental issues.

• Specific mechanisms are developed to facilitate compliance with regulations or policy requirements. For example, the EU Farm Advisory System was originally meant to help with cross-compliance conditions; US conservation policies include funding for technical advice on how to comply.

• As a result of efforts to reduce costs and increase coverage, public systems increasingly provide advice to groups of farmers, while farmers are expected to pay for individual advice. This often goes along with a reduction in the number of advisors as mentioned in the Australian and Estonian reviews.

• ICTs are increasingly used to transfer knowledge and information, thus allowing for better inclusion of remote farmers.

• It is difficult to interpret changes in public funding over time, as it may reflect the growing importance of a healthy private sector or a decrease in access.

• New intermediary, private actors have emerged following the privatisation of government services (in the Netherlands), or the introduction of competition (in Turkey), but also to help with new technologies. For example, more efficient and cheaper digital technologies have prompted the creation of knowledge intermediaries to help farmers benefit from digital services).

• Some countries focus public provision on public good aspects, e.g. poor farmers (for example in Brazil and Colombia), and policy and environmental aspects (for example in Estonia and the United States).

• Some countries pay farmers to access advisory services, rather than subsidise the service itself.

• Some countries try to reach out farmers who do not use advisory services, e.g. in Brazil and Colombia.

• The main role for the government is in the governance of the system to ensure provision is adequate, and all farmers have access to a competitive supply of advisory services, covering both productivity and sustainability aspects, and type of advice (technology, management, policy or marketing). Governments should in particular:

  • Set qualitative and quantitative objectives, and guide the evaluation of performance.

  • Ensure extension officers are well-trained (certification) and keep informed about the most recent knowledge via re-training or participation in innovation networks.

  • Facilitate networking and exchanges of knowledge between advisors and with other actors.

  • Facilitate the use of most advanced technologies to communicate.

  • Focus public funding on public good aspects through supporting services or farmers access to services.

  • Explore innovative ways to reach farmers who are out of the system, for example by making participation a condition for receiving support.

Farm advisory systems are very different across countries and often include many public and private providers, enabling farmers to choose. However, this makes it difficult to share experience and allow evaluation. Few countries mention evaluating advisory services, although the evaluation of government spending may follow the framework of general policy evaluation in some countries. For example, policy evaluation in Estonia led the government to target public services to farmers with lack of access to private ones. Moreover, ad hoc research studies have addressed the effectiveness of advisory systems, in particular in emerging economies. A general conclusion is that further analysis is needed to better understand farmers' needs and main barriers to adoption.

Given enabling market and policy incentives, other mechanisms to facilitate innovation in food and agriculture, include participation in networks, co-funding in public-private projects, investment support in farms and firms.

International research co-operation

International co-operation on agricultural research and development offers universal benefits. While this is generally true given the public good nature of many innovations in agriculture, it is particularly the case with global challenges — as in the case of responding to climate change — and when initial investments are exceptionally high. The benefits of international co-operation for national systems stem from the specialisation it allows and from international spill-overs. In countries with limited research capacity, scarce resources could then focus on better taking into account local specificities.

The importance of bilateral and multilateral co-operation in R&D and technology transfer is well recognised and various mechanisms, which are not necessarily agriculture-specific, facilitate this co-operation. They include exchange of students and staff, which are encouraged in many countries (e.g. Canada); and co-financing of and participation in international projects, initiatives and networks. Brazil developed an interesting exchange mechanism to facilitate international collaboration (Box 5.6).

Box 5.6. The Brazilian Virtual Laboratories Programme (Labex)

Embrapa in Brazil created Labex (Virtual Laboratories Programme) to promote opportunities for institutional co-operation in agricultural research and to monitor scientific advances, trends, and activities of interest to agribusiness in partner countries. Embrapa selects and sends senior researchers to develop strategic research for Brazil in partnership with R&D centres of excellence in agricultural research. Since 1998, Embrapa has opened virtual laboratories in the United States, Europe (France, England, the Netherlands and Germany), Korea, China and Japan.

The programme also provides for the reverse route: at the so-called “reverse Labex”, researchers from partner international institutions are established in Embrapa's research centres to develop projects of mutual interest.

Source: OECD (2015f), update in www.embrapa.br/en/web/portal/embrapa-labex.

An important part of international collaboration for agricultural R&D concerns agricultural development, for example in the CGIAR system (www.cgiar.org/), or the Global Forum on Agricultural Research (GFAR – www.egfar.org). In this area, international efforts also aim to develop capacity for agricultural innovation in developing countries.10

Reviewed countries also participate in a number of international initiatives focusing on global challenges, in particular climate change.11 These include G20 initiatives for collaboration on agricultural research, for example the Wheat Research Initiative (IRIWI – www.wheatinitiative.org); the Group on Earth Observations Global Agricultural Monitoring Initiative (GEOGLAM – www.geoglam.org/index.php/en/).

EU innovation policy aims to bring together researchers from different EU Member States. The Standing Committee on Agricultural Research (SCAR) plays a major role in the co‑ordination of agricultural research efforts across the European Research Area (ERA), including by establishing Collaborative and Strategic Working Groups to define common priorities. As part of the ERA, successive EU Framework Programmes and current Horizon 2020 fund multi-country teams, selecting projects on a competitive basis. In addition, different initiatives also facilitate cross-country collaboration. They include ERA-nets, Joint Programming Initiatives (JPIs) and more recently European Innovation Partnerships (EIPs). The reviewed EU Member States participate jn various ERA-nets (e.g. on pest and disease research), in the JPI on Agriculture, Food Security and Climate Change (FACCE), and the EIP on Agricultural Productivity and Sustainability. Moreover, fostering international co-operation in research and innovation is a strategic priority for the European Union, and participation of researchers from third countries in EU-funded projects and initiatives has gradually increased over time.

In most countries, the share of scientific publications on agro-food research including at least a foreign co-author is above 30%, and reaches levels above 60% in the Netherlands, Sweden and Switzerland, reflecting both high research capacity and focus on collaborative work, in particular at the EU level (Figure 5.5). The share of patents with foreign co-inventors is the highest Switzerland (54%) and Argentina (45%), and is between 20% and 30% in a majority of reviewed countries. The relatively modest share of outputs from co-operation with foreign researchers in the United States reflects the large share of national R&D by international standards, rather than the small level of co-operation with other countries.

Few country reviews mention significant obstacle to international co-operation, although the mandate to focus on national issues may limit the scope for it. There are also concerns in some countries that restrictions on immigration may limit exchanges of researchers and students, identified as an efficient mechanism to facilitate cross-country co-operation.

Figure 5.5. International co-operation in agro-food R&D, 2007-12

1. Publications in scientific journals. OECD average excludes Lithuania. Data refer to agro-food outputs for 2007-12.

2. Patents filed under the Patent Co-operation Treaty (PCT). Data refer to agri-food outputs for 2006-11.

Source: OECD (2014), OECD Patent Database https://stats.oecd.org/ (accessed February 2014); SCImago. (2007), SJR — SCImago Journal & Country Rank, www.scimagojr.com (accessed 19 March 2014).

Statlink

Summary of innovation policy indicators

Table 5.4 recapitulates key indicators that characterise Agricultural Innovation Systems and innovation policy approaches in reviewed countries.

Table 5.4. Summary of the main characteristics of Agricultural Innovation Systems and innovation policy approaches

	Governance: Evaluation practices	Agricultural research intensity	Food industry research intensity	IPR	R&D steering	Research outputs		Research output with foreign co-author		Extension system
	1: ad hoc; 2: project and staff 3: 2+ institution level 4. 2+wrt policy objectives 5: 3+4 6: 3+4+RoI 7: 6+ other IA	Public expenditure on agricultural R&D as a % of agricultural GVA1	Food industry expenditure on R&D as a % of Food manufacturing GVA2	1 to 7 (highest)	Share of project-based funding in total funding for agricultural R&D: 1: 0-20%; 2: 20-40%; 3: 40-60%; 4: 60-80%; 5: 80-100%	Number of agro-food patents relative to the sector's GVA (USD billion)	Number of agro-food publications relative to the sector's GVA (USD billion)	As a % of all agri-food patents	As a % of all agro-food publications		1. Diverse and interactive; 2. Strong public extension services; 3. Strong role of farmer organisations; 4. Dual-unequal
Argentina	1 and 3	0.6	..	3.6	1			44.3	37.5		4
Australia	7	1.5	..	5.8	3			23.1	47.3		1
Brazil	3 and 6	1.8	..	4.1	1	0.4	50	29.7	22.3		4
Canada	4	1.9	0.5	5.9	2	4.0	157	29.7	48.9		1
China	..	0.6	..	4.3	1			21.8	23.6		2
Colombia	4	0.8	..	4.2	4			29.4	54.5		4
Estonia	4	1.4	0.8	5.5	4	3.8	170	30.6	47.3		2
Japan	4	1.8	1.7	5.9	1	7.0	51	5.2	31.5		2
Korea	4	3.0	2.7	4.4	1	6.9	60	5.8	31.4		2
Latvia	4	..	0.2	4.2	2	0.3	24	16.7	46.9		3
Netherlands	5	0.9	2.7	6.2	5	11.7	112	27.1	65.1		1
Sweden	3	0.9	1.0	6.1	3	11.2	223	26.9	62.9		1
Switzerland	2	2.2	0.5	6.5		15.6	196	53.7	68.1
Turkey	2	0.2	0.2	3.7	2			27.9	18.6		2
United States	7	1.4	2.7	5.9	4	8.8	94	14.3	36.4		2

Notes: ..: Not available; GVA: Gross Value Added; 1. GBARD; 2. Business Expenditure on R&D (BERD) is the measure of intramural R&D expenditures within the business enterprise sector (regardless the sources of R&D funds).

Main knowledge gaps

Country reviews generally contain a good description of the agricultural innovation actors and funders, and the governance structure. It is however difficult to assess the functioning of the mechanisms to set priorities, co‑ordinate with other policies and sectors, consult stakeholders. The nature and scope of evaluation procedures are also difficult to assess.

Overall, information needed to monitor and evaluate R&D and innovation in food and agriculture, and to assess the performance of Agricultural Innovation Systems is not sufficient.

The most surprising finding is the difficulty to track funding of agricultural R&D over time and across countries. Despite the existence of international standards for reporting R&D expenditure, reporting at sector level is often incomplete in the OECD database. The OECD database includes different indicators of expenditures on R&D, mainly GBARD, Government Expenditure on R&D (GERD) and BERD, and its components. If coverage is quite good at the national level, there are multiple gaps and discrepancies at the sector level. GBARD for agriculture is the most widely available and most timely as it originates from budget sources, but it only captures public expenditures budgeted for agriculture. However, some countries increasingly use general funding mechanisms to finance agricultural research. GERD and BERD come from surveys and the time series are sometimes incomplete or dated.

GERD for agricultural sciences is mostly available for two sectors of performance: government and higher education, and includes both public and private funding of research performed in these sectors. Public and private funding of R&D for agricultural sciences in the private sector is often missing. Moreover important countries like the United States do not report any data according to this classification. Series for GERD by sector of performance for agriculture as an economic objectives are shorter and mainly incomplete. GERD by funder is not available.

Information on private funding of research on food and agriculture is incomplete and often lacking in smaller countries and for smaller companies. In most countries, the diversity of research funding channels, and of research organisations makes it difficult to track public funding to its end-use. As a result, the OECD database does not allow tracing sources of funding to sectors of performance like in Figure 7.4 of the US review (OECD, 2016c). BERD in food processing companies has good coverage but not BERD for agriculture.

Quantitative information on funding mechanisms is not available in all countries, and assumptions had to be made to fill the fifth column in Table 5.4. Similarly, there is little information on the distribution of funding by level of research from fundamental to applied, by commodity sector or by issue.

As part of innovation policy, governments support R&D and innovation, through various mechanisms and programmes. Some country reviews contain detailed information about these, but not much on whether agriculture and food companies benefit from them.

Few country reviews report on the results of national evaluation of agricultural innovation systems (Mainly Australia, the Netherlands and the United States). All reviews include indicators of research outputs (bibliometrics and patents) that come from international databases, but they do not necessarily cover all types of research activities. More efforts are needed to measure research outputs, but also to build more meaningful indicators for cross-country comparison. Country reviews include relative indicators of the share of agri-food patents or publications in the national total, which indicate the relative specialisation of the research system, to compare to the share of the sector in the economy. They also include the country’s contribution to world totals, which are particularly interesting for large countries. To evaluate research efficiency, research outputs would need to be related to some indicator of size or efforts, such as the number of researcher or the expenditure on agricultural research, but coverage often differ. Patent and publication numbers cover both public and private sector, as well as agriculture and food processing, while research effort covers mainly researchers and research expenditure in public organisations. To facilitate cross-country comparison of availability of research output for the sector, Table 5.4 presents the number of agro-food patents and publications relative to the sector's gross value added.

Evaluation of research impacts poses methodological challenges, in particular in terms of attribution, evaluation of longer term impacts and non-market impacts (Alston, 2010; Joly et al., 2016). Some countries are developing procedures and guideline, and efforts are made at international level in this area (OECD, FAO), but more concerted efforts are needed.

Except in ad hoc surveys, most countries have little information on farm-level adoption of innovation and use of advisory services. This makes it difficult to identify the factors that facilitate adoption and the impacts of adoption on farm performance. More information is available on firm-level innovation, as part of EU industry and innovation surveys, mainly for the food and drink processing industry. While the role of advisory systems in the adoption of innovation is recognised, there is little evidence available on the performance of specific extension systems and their ability to reach out the farmers most in need to improve their productivity and sustainability performance.

Recommendations to improve the agricultural innovation system

This section consolidates recommendations to strengthen agricultural innovation systems made in country reviews, which are included in the country notes of Annex B.

Improving governance of agricultural innovation systems is key

• Establish a longer-term strategy for agricultural innovation to guide operational objectives and associated expenditures, taking into account long-term challenges such as climate change, as well as societal demand.

  • Ensure coherence between innovation and growth strategies;

  • Improve the integration of agricultural innovation objectives in government-wide innovation strategy;

  • Improve involvement of stakeholders in the definition of objectives, starting at an early stage of the process; and

  • Include measurable targets in the definition of objectives; and monitor progress towards objectives.

• Improve co‑ordination between research organisations, public and private and at the national and sub-national levels:

  • If missing, create a specific national institution (e.g. a national council) to co‑ordinate objectives and monitor policy, and ensure continuity in programming.

  • Clarify mandates of organisations to avoid duplication or under-supply in one area (for example between research and knowledge transfer).

• Develop coherent evaluation procedures at different levels (researchers, projects, institutions, system) that include some independent evaluation and cover a wide range of indicators of efforts, outputs and impacts to allow for future improvements. Ensure evaluation criteria are consistent between levels and with objectives.

Strengthen linkages within the national innovation system to increase efficiency and responsiveness to needs

• Facilitate linkages within the agricultural innovation system (between research, advisory services, education, government, farmers and agri-food companies) and with other experts and stakeholders.

• Promote and enable research co-operation across sectors, so that food and agriculture benefit from advances in other sectors and generic research, such as genetics, digital technologies.

• Remove institutional constraints to public research organisations to engage in co-operation activities with the private sector.

• Facilitate the organisation of producers and industry to enable them to contribute more effectively to the agricultural innovation system.

• Facilitate the emergence Public-Private Partnerships (PPPs) for research and innovation, when they bring additional benefits. Provide guidelines for successful PPPs, and in particular, ensure objectives are shared, institutional arrangements are clear, including the sharing of costs and benefits between partners, which should be commensurate.

• Create and support poles of competitiveness, or poles of excellence to facilitate co-operation.

• Explore further opportunities to share public infrastructure with the private sector.

• Identify areas where local companies and researchers could collaborate to develop local or niche products and innovation.

• Support the functioning of local, national and international networks for innovation and the participation of researchers and stakeholders in these networks.

• Strengthen the links between R&D and technical assistance, for example by adding technology transfer component to research projects, or by encouraging networking between researchers, advisors and producers.

Simplify research programming to improve effectiveness and transparency

• Simplify the programming of R&D and innovation funding, and provide clear information to improve access. For example, streamline funding programmes and build a single platform that informs on all the sources of available government funding. When relevant, include information on sub-national sources.

• Review the efficiency of research funding mechanisms to ensure higher impact. Consider greater use of incentives that incentivise transdisciplinary and system-based approaches, and wider stakeholder involvement that increases relevance. Explore ways to generate new (breaking through) ideas to overcome current constraints, for example through demand-driven funding mechanisms.

• Explore ways to generate new (breaking through) ideas to overcome current constraints, for example through demand-driven funding mechanisms.

Focus public funding of agricultural R&D in areas with public good aspects to improve complementarity with other efforts

• Provide stable funds for knowledge infrastructure, including knowledge technologies, institutions, networks and databanks, and long-term, large scope projects to maintain or strengthen the capacity of public research to meet public objectives and collaborate with private and international partners.

• Improve complementarity of public research funding:

  • with private efforts, by focusing public funding to areas that are not covered by, or complement, private efforts (including by agri-food companies and value-chain organisations), in particular areas with public goods aspects such as management of natural resource use, low carbon technology, and resilience to large-scale risks, and global food security.

  • at the national, sub-national and international levels. In EU Member States, improve consistency between EU and national rules and objectives to facilitate participation in EU programmes.

• Dedicated specific funds for policy-relevant research, i.e. research that generates information needed to improve policies.

• In country where research is organised by commodity sector, create cross-sector thematic areas and projects, including environment, or broaden the scope and membership of existing commodity research systems.

• Explore ways to generate new, break-through solutions to current and future challenges.

Strengthen private contribution to R&D and innovation for food and agriculture to increase impact

• Enhance the involvement of processing industries and retailers in innovation, by making them an integral part of the system, from the priority setting stage to the financing and commercialisation of innovation stages.

• Evaluate programmes that support innovation in private companies to ensure they are efficient and reach intended beneficiaries. In particular, monitor whether their reach agri-food companies. If private companies have access to tax incentives for R&D, evaluate the system to ensure it stimulates additional research activities.

• Strengthen and harness the capacity of private companies to participate in research partnerships via project funding, support to networking, training activities, and effective Intellectual Property Rights (IPR) protection.

  • Ensure appropriate protection of IPR, and improve enforcement when needed, to attract private funding, while allowing for further use for research purpose, as with Plant Breeders’ Right.

• Explore alternative sources of funding for research and innovation:

  • This could include farmers’ contributions, and revenues from royalties or Intellectual Property.

  • Investigate the demand and supply for venture capital for agri-food companies, and identify possible role for the government to ease constraints.

  • Lower barriers to FDI in agricultural R&D if they exist.

Facilitate international R&D co-operation

• Explore opportunities for public research to engage in bilateral, regional and multilateral co-operation in R&D and technology transfer

• Remove institutional constraints to public research organisations to hire foreign researchers of trainees

• Providing positive incentives such as support to student and staff exchange, sharing of equipment and laboratories.

Strengthen farm advisory systems to facilitate adoption

• There is no one size fits all or preferred model for farm advisory systems.

• Encourage a diverse supply of relevant advice from diverse public and private suppliers.

• But ensure needs are met:

  • Review current systems, and identify needs and gaps.

  • Ensure advisory services cover technical, financial and organisational aspects, and sustainability improvements.

  • And advice is accessible for all types of farmers, using new (digital) technologies.

• Focus public role on services that the private sector eventually under-provides:

  • Target the needs of small, semi-subsistence farmers to broaden their opportunities.

  • Strengthen environmental performance by providing targeted advice on sustainable technologies and practices, and use experience to better understand issues and needs.

• Explore the scope for including support to technical assistance and research projects within agri-environmental policies, in case of under-provision.

• Facilitate the sharing of experiences through networking, the development of databases.

• Ensure advisors have up-to date knowledge, possibly through certification, and facilitate continuous training.

Facilitate knowledge sharing and information dissemination

• Continue developing information systems to guide policies, research and innovation, and facilitate knowledge sharing.

  • Include market intelligence (big data) and research results.

  • Monitor innovation adoption in form and farm surveys, surveys to generate data on innovation and better understand motivations and barriers to adoption.

  • Monitor environmental performance in surveys.

  • Use innovative methods to reduce collection costs and improve farm and firm participation, drawing on experience from other countries.

  ‒ Develop indicators and tools to evaluate the performance of the agricultural innovation systems in general, and innovation policy regularly, taking longer-term effects into account.

• Promote the integration of research data and sharing of experience at the international level.

• Improve public understanding of the importance of innovation in food and agriculture, in the sector and society, and build trust in science through increased transparency and education.

References

Acil Allen Consulting (2014), CSIRO’s Impact and Value. An Independent Evaluation, Melbourne. www.acilallen.com.au/cms_files/ACILAllen_CSIROAssessment_2014.pdf.

Alston, J. (2010), “The Benefits from Agricultural Research and Development, Innovation, and Productivity Growth”, OECD Food, Agriculture and Fisheries Papers, No. 31, OECD Publishing, Paris, https://dx.doi.org/10.1787/5km91nfsnkwg-en.

ASTI (2017), Agricultural Science and Technology Indicators (database), www.asti.cgiar.org/data (accessed March 2018).

Beintema, N., G. Stads, K. Fuglie, and P. Heisey (2012), ASTI Global Assessment of Agricultural R&D Spending, International Food Policy Research Institute, Washington, DC.

Google ScholarCSIRO (2015), Impact Evaluation Guide, CSIRO, Performance & Evaluation Unit, www.csiro.au/impact.

Day-Rubenstein, K. and K. Fuglie (2011), “Research and Development in the Food Manufacturing Industry”, Chapter 9 in Fuglie et al. (2011), Research Investments and Market Structure in the Food Processing, Agriculture Input and Biofuel Industries Worldwide, Economic Research Report 130, Economic Research Service, US Department of Agriculture, Washington, DC, www.ers.usda.gov/publications/pub-details/?pubid=44954.

Fuglie, K., P. Heisey, J. King, C. Pray, K. Day-Rubenstein, D. Schimmelpfennig, S-L. Wang and R. Karmarkar-Deshmukh (2011), Research Investments and Market Structure in the Food Processing, Agriculture Input and Biofuel Industries Worldwide, Economic Research Report 130, Economic Research Service, US Department of Agriculture, Washington, DC, www.ers.usda.gov/publications/pub-details/?pubid=44954.

Fuglie, K. (2016), “The growing role of the private sector in agricultural research and development world-wide”, Global Food Security, Vol. 10, pp. 29-38, https://doi.org/10.1016/j.gfs.2016.07.005.

Joly, P., et al. (2016), “Agricultural research impact assessment: Issues, methods and challenges”, OECD Food, Agriculture and Fisheries Papers, No. 98, OECD Publishing, Paris, https://dx.doi.org/10.1787/5339e165-en.

MAFF (2016), Field for Knowledge Integration and Innovation: Organization and Evolution Since Fiscal Year 2016, Ministry of Agriculture, Forestry and Fisheries, Tokyo, www.knowledge.maff.go.jp/uploads/0f46f5c7b37748f264f227c4073ffd134b453ba3.pdf.

Moreddu, C. (2016), “Public-Private Partnerships for Agricultural Innovation: Lessons from Recent Experiences”, OECD Food, Agriculture and Fisheries Papers, No. 92, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jm55j9p9rmx-en.

OECD (2019a), OECD statistics [Research and Development, OECD National Accounts], https://stats.oecd.org/ (accessed January 2019).

OECD (2019b), Digital Opportunities for Better Agricultural Policies, OECD Publishing, Paris, https://doi.org/10.1787/571a0812-en.

OECD (2018a), Agricultural Policy Monitoring and Evaluation 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/22217371.

OECD (2018b), National Accounts (database), [Value added and its components by activity, ISIC rev4], Research and Development Statistics (database), [Business enterprise R-D expenditure by industry (ISIC 4)]; MSTI Main Science and Technology Indicators (database), [BERD as a percentage of GDP], https://stats.oecd.org/.

OECD (2018c), Innovation, Agricultural Productivity and Sustainability in Estonia, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264288744-en.

OECD (2018d), Innovation, Agricultural Productivity and Sustainability in Sweden, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/9789264085268-en.

OECD (2017), R&D Tax Incentive Indicators, https://oe.cd/rdtax.

OECD (2016), Innovation, Agricultural Productivity and Sustainability in the United States, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264264120-en.

OECD (2015a), Innovation, Agricultural Productivity and Sustainability in Canada, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264238541-en.

OECD (2015b), Innovation, Agricultural Productivity and Sustainability in the Netherlands, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264238473-en.

OECD (2015c), Innovation, Agricultural Productivity and Sustainability in Australia, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264238367-en.

OECD (2015d), OECD Review of Agricultural Policies: Colombia 2015, OECD Review of Agricultural Policies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264227644-en.

OECD (2015e), Fostering Green Growth in Agriculture: The Role of Training, Advisory Services and Extension Initiatives, OECD Green Growth Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264232198-en.

OECD (2015f), Innovation, Agricultural Productivity and Sustainability in Brazil, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264237056-en.

OECD (2014), OECD Patent Database, https://stats.oecd.org/.

OECD (2013), Agricultural Innovation Systems: A Framework for Analysing the Role of the Government, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264200593-en.

Pray, C.E. and K.O. Fuglie (2015), “Agricultural Research by the Private Sector”, Annual Review of Resource Economics, Vol. 7, pp. 399–424, https://doi.org/10.1146/annurev-resource-100814-125115.

Productivity Commission (2011), “Rural Research and Development Corporations”, Report No 52, Final Inquiry Report, Canberra, www.pc.gov.au/inquiries/completed/rural-research.

SCImago (2007), SJR — SCImago Journal & Country Rank, www.scimagojr.com (accessed 19 March 2014).

Toole, A.A. (2014), “Emerging Opportunities for Public-Private Partnerships in the United States”, presentation to the 6th meeting of the OECD Food Chain Analysis Network on PPPs for agricultural innovation, 13-14 October 2014, Paris, www.oecd.org/site/agrfcn/meetings/6th-oecd-food-chainanalysis-network-meeting-october-2014.htm.

World Economic Forum (2017), The Global Competitiveness Report 2017-2018: Full data Edition, Geneva 2017, http://reports.weforum.org/global-competitiveness-index-2017-2018.

Notes

← 1.  TÜBITAK is the Scientific and Technological Research Council of Turkey. Its mandate is to guide, co‑ordinate and fund national science, technology and innovation. It also performs research in strategic R&D areas, including agro-food research in the Food institute and the Genetic Engineering and Biotechnology Institute.

← 2.  Government budget allocations for R&D (GBARD) are more widely and timely available across countries than Government Expenditure on R&D (GERD). However, it provides only a partial indicator of investment in public agricultural research, since it refers to research funding instruments dedicated specifically to agriculture. In some countries, however, agricultural research funding comes through more general competitive funding programmes that are not captured in “agricultural” GBARD.

← 3.  Updating the database built used in Fuglie et al. (2011), Fuglie (2016) estimates that global private agricultural R&D spending tripled in nominal terms during 1990-2014, while private spending for food also tripled during 1990-2012. The database on global private agricultural R&D tracks agriculture-related R&D expenditure by major companies conducting research in this area, adding an allowance for R&D spending by small and mid-size firms. Companies are classified into seven farm input sectors. To allocate expenditures by countries, private R&D is assumed to take place in the country where the corporate headquarters is located, or where sales are made. The database covers the period 1990-2014. R&D expenditures by the food industry cover the period 1990-2012 and draws on OECD data (Business expenditure on R&D, BERD) for OECD and other countries covered in the OECD database on R&D statistics, and for others, from national estimates, where available. If not, expenditures are assumed to be zero (Fuglie, 2016).

← 4.  Fuglie et al. (2011) found that the largest 5-10 companies in each sub-sector of agriculture-related activities accounted for 80% or more of total R&D in the sub-sector. Fuglie (2016) found that the 23 top-tier companies accounted for over 70% of global private expenditure on agricultural R&D in 2014.

← 5.  Adding the global estimate of public expenditure on agricultural R&D from Beintema et al. (2012) to the estimate for private expenditure in Fuglie (2016), the share of the latter in the total is about 30%.

← 6.  OECD seeds schemes web site: www.oecd.org/tad/code/abouttheoecdseedschemes.htm.

← 7.  Pray and Fuglie (2015) found that technology policy (support, IPR, and access to national market) can have a significant influence on private agricultural R&D spending in emerging economies, including by foreign companies. They analyse the experiences of Brazil, China and India, contrasting the approaches and identifying the respective contribution of multinational versus national companies.

← 8.  See Box 6.1 in OECD (2015a).

← 9.  Pull-mechanisms reward successful innovations ex post, as compared to push mechanisms, which fund potential innovations ex ante (see Box 6.3 in OECD, 2013).

← 10.  For example the Tropical Agricultural Platform (TAP) (www.fao.org/in-action/tropical-agriculture-platform/background/en). Developed in this context, TAPipedia is an information sharing system designed to enhance knowledge exchange in support of Capacity Development (CD) for Agricultural Innovation Systems (AIS). It aims to be a global information system for good CD practices, innovation outputs, success stories and lesson learned.

← 11.  For example the International Panel on Climate Change (www.ipcc.ch/) and the Global Research Alliance on Agricultural Greenhouse Gases.