Bio-based and Applied Economics 4(3): 279-300, 2015 ISSN 2280-6180 (print) © Firenze University Press ISSN 2280-6172 (online) www.fupress.com/bae DOI: 10.13128/BAE-17555 Review Article Research and innovation in agriculture: beyond productivity? DaviDe viaggi Department of Agricultural Sciences (DipSA), University of Bologna, Viale Fanin, 50 - 40127 Bologna (ITALY) Date of submission: October 5th, 2015: accepted November 15th, 2015 Abstract. Studies on the effects of research and innovation in agriculture have been largely characterised by efforts to make a connection between expenditure and pro- ductivity. A number of issues have challenged the ability of productivity to measure the effects of research, namely, in recent years, increasing efforts towards improving the environmental performance of the farming sector. Besides environmental con- cerns, however, a number of recent concepts have emerged that are shaping the cur- rent research and policy agenda and which could result in a revision of the productiv- ity concepts used to evaluate research impacts. The objective of this paper is to discuss these issues and their implications for studies on the impact of research and innova- tion. We address, in particular, the following issues: a) the development of the of bio- economy and related concepts such as the circular economy, resource efficiency and bio-refinery; b) the connection with entrepreneurship and eco-innovation; c) chang- ing tools in research assessment, in particular the widespread use of Life Cycle Assess- ment (LCA); and d) the evolving concepts of sustainability and ecosystem services. We argue that while the traditional notion of productivity, intended as output/input ratio, maintains (and may be strengthens) its role on the aggregate, a more analytical interpretation of the pathways towards research impacts is needed, as well as a broad- ened view of productivity and its determinants. Keywords. Agriculture, research, innovation, productivity, bio-refinery, circular economy, bioeconomy JEL Codes. Q16 1. Introduction and objectives The link between research, innovation and productivity has gained growing atten- tion in the last decade. The productive focus of agriculture in the mid-twentieth century appeared rather straightforward. Yet the environmental crisis of the late twentieth century brought a wider variety of issues to the forefront. At the outset of the twenty-first cen- Corresponding author: davide.viaggi@unibo.it. 280 D. Viaggi tury, the dilemma in EU agriculture appeared to be whether to move towards more tra- ditional, organic farming or to modern biotechnology-based agriculture (De Wilt et al., 2001). Now, however, the scenario is more complex. On the one hand, different agricul- tural technologies and sustainability pathways co-exist (worldwide at least); on the other hand, attention has moved from choosing between existing technologies to a focus on new technological developments through innovation. At the same time, the concept of sustainability is changing: initially it was focused mainly on environmental effects, then enlarged to economic and social sustainability and resource efficiency. The recent decade has refocused attention on feeding the word while preserving resources for future gen- erations, which is the “mantra” of the current debate and recent events (see EXPO 2015), with a clear renewed attention on productivity. Policy has followed and promoted this change in focus. An example is the evolu- tion of the Common Agricultural Policy (CAP) from providing incentives to production through market support, to focusing more on income support, to a mix of sustainability and income support, and to the present aggregate of productivity, resource efficiency and climate change concerns. The relationship between research, innovation and productivity has taken on a grow- ing policy focus in the EU (in particular with the EU2020 strategy), with an increasing focus on competitiveness, innovation and resource efficiency. This has also affected subse- quent policy implementation, including the CAP 2014-2020. In this programming period, and in particular in the rural development component, innovation has a prominent role, including new connections with research, especially through the new instrument repre- sented by the European Innovation Partnership (EIP). The focus of agricultural research in the last two centuries has followed the needs expressed by society and policy, with deep changes over time (Spierz, 2014). In recent years, increasing attention to public good production by agriculture (environmental, working quality etc.) has affected both the focus of research and the breadth of impacts that would be necessary to measure to account for these effects. A general definition of productivity is the ability of production factors to produce output (Latruffe, 2010). This can be measured through a ratio between output and input. However, a wide body of literature has treated productivity measures in agriculture addressing both conceptual and practical issues in implementing this concept. The liter- ature includes simple measures of partial productivity (e.g. relating output to individual inputs as well as measures accounting for more than one input (e.g. multiple factor pro- ductivity). A more comprehensive measure of productivity is Total Factor Productivity (TFP), which is a ratio of the (monetary) aggregation of all outputs and the aggregation of all inputs. This concept often follows a time series approach to account for productivity improvements based on changes in TFP (Coelli, 2005; Latruffe, 2010). Fuglie (2015) shows that TFP has a major role in accounting for the growth of agricultural production over time, though with a differing weight in different time periods and in different regions. Productivity indexes have been discussed and extended to account for a number of issues, including the growing variety of inputs and outputs related to environmental and resourc- es concerns. Several multi-output, resource-saving, or dynamic specifications are now attempting to tackle these issues. This has led, among others, to proposing extensions such as environmentally adjusted TFP or even green TFP (e.g. Chen and Golley, 2014). 281Research and innovation in agriculture: beyond productivity? Besides measuring productivity, the economic literature has attempted to investigate the factors affecting productivity and its changes. This has been done largely through econometric models using productivity measures as dependent variables and various influencing factors as explanatory variables. Among such factors, research and innova- tion have a prominent role. In particular, the research to date has focused on seeking to make a connection between research expenditure and productivity (Alston et al., 2010, 2011; Wang et al., 2013). Research and innovation are indeed potentially key explanatory variables of changes in productivity over time. Through the provision of new knowledge (increasing the knowledge stock), embedding this knowledge in new technologies (enlarg- ing the technology set) and the diffusion of innovation in the economic system, research is expected to directly affect the relationship between input and output and hence improve productivity. In this way, the role of research and innovation can be seen as a way of escaping (or at least modifying) trade-offs among goods, given limited resource availabil- ity, by enhancing technical possibilities. Yet as research is costly, this implies a trade-off between present consumption and investment in technology development. Methods used to assess this connection may utilise econometric techniques on large data sets of (relatively simplified) information, as well as case study type approaches (Alston et al., 1995; Gaunand et al., 2015). Altogether, the link between research, inno- vation and productivity is conceptually straightforward, but much less easy to streamline in practice. This is partly due to technical issues, such the practical difficulties in prov- ing numerically the linkage between research and productivity change, in part due to data limitations. Further conceptual difficulties are due to the complexity of the system con- necting research, innovation and productivity. This is underscored by, among others, the long-standing debate about the difficulty in matching demand and supply of innovation, the on-going discussion regarding the functioning of knowledge and innovation systems, and the growing discussion on the multiple directions (functions and objectives) of agri- culture and related research. There are also potential counterintuitive effects of research, such as the negative impacts of highly codified public research on local enterprises (Mai- etta, 2015). Making this linkage has also become more difficult over time due to the increased variety of technology pathways (conventional, integrated, organic and more) used in agri- culture and of the goods (private, public, intermediate) produced by agriculture. These issues have been exacerbated in recent years by the emerging focus on the bioeconomy and renewed attention devoted to resource efficiency and productivity. The concepts related to the bioeconomy emphasise the complexity of technical and economic linkag- es that can transmit the effects of research (Viaggi et al., 2012; Swinnen and Weersink, 2013). Finally, worldwide trends, including the opening of various economies and climate change, have given additional weight to uncertainty and dynamics, with a special empha- sis on non-linear and “non-trend” dynamics and fundamental uncertainty in the outcome and relevance of any technology change and policy action. It is now rather well established that the assessment of the impacts of research (espe- cially public research) should deal with multiple dimensions, covering not only economic, but also environmental, health, social, and policy impacts. Furthermore, it is accepted that innovation results from the activities and interactions of multiple actors. Understanding the impact of a research activity/organisation increasingly requires paying attention to the 282 D. Viaggi intermediaries as well as the beneficiaries (i.e. the innovation network) of the research results (Gaunand et al., 2015). While some of the issues above have already been covered extensively in the literature, others are rather new and the scientific community is still in the process of identifying the challenges ahead. The objective of this paper is to discuss the link between research, innovation and productivity in agriculture in light of selected emerging economic and sustainability con- cepts, mainly, but not exclusively, linked to the development of the bioeconomy. The paper addresses, in particular, the following issues in relation to the link between research and productivity: a) the concept of ‘bioeconomy’ and related concepts such as the circular economy, resource efficiency and bio-refinery; b) the connection with entrepre- neurship and eco-innovation; c) changing tools in research assessment, in particular the widespread use of Life Cycle Assessment (LCA); and d) the evolving concepts of sustain- ability and ecosystem services. In relation to these fields, this paper attempts to identify implications for the concept of productivity and for the understanding of the linkage between research and productivi- ty, in particular seeking to identify developments that may challenge the current notion of productivity. This reasoning is then used to develop implications for future research. The paper does not intend to focus specifically on the measurement of productivity (though this is clearly an important issue in this context and is implicitly or explicitly addressed at several points in the paper). Rather, it focuses on the connection between objectives/ mechanisms and the effects of research. Moreover, while recognising that the issue has relevance for all scales of analysis, the focus of this paper is implicitly more on the micro to meso level and closer to the research level, compared with many productivity-oriented studies. In addition, it tends to be more on the threshold between decision support and assessment methods, largely because the paper seeks significant inspiration from trends in evaluation practice and decision making concerning research and innovation funding. We do not restrict attention to a specific field of research; while at the same time not excluding research in the social sciences, however, the paper focuses more on “classical” technological research. The paper is organised in three main sections beyond this one. In the next Section (2), we review the literature on the four issues listed above and their implications for trac- ing the effects of research on productivity; in Section 3 we discuss the implications of the findings from the literature for future research and evaluation practices. Concluding remarks are provided in Section 4. 2. Research and productivity: selected issues 2.1 Changing paradigms in economic organisation: bioeconomy, circular economies and bio- refinery The concept of the bioeconomy, based on the sustainable exploitation of biological resources, has emerged as a key strategy in many countries to meet human needs while taking into account resource efficiency requirements. After having proposed several dif- 283Research and innovation in agriculture: beyond productivity? ferent definitions in recent years, the EU Communication on the Bioeconomy and the accompanying working document (European Commission, 2012a; European Commis- sion, 2012b) defines the bioeconomy as encompassing “the production of renewable bio- logical resources and their conversion into food, feed, bio-based products and bioenergy. It includes agriculture, forestry, fisheries, food and pulp and paper production, as well as parts of chemical, biotechnological and energy industries. Its sectors have a strong innova- tion potential due to their use of a wide range of sciences (life sciences, agronomy, ecolo- gy, food science and social sciences), enabling and industrial technologies (biotechnology, nanotechnology, information and communication technologies (ICT), and engineering), and local and tacit knowledge.” Based on this delimitation, the EU bioeconomy accounts for an annual turnover of more than 2 billion euro (of which almost half from the food sector) and more than 20 million employees. Besides its economic importance and potential, the bioeconomy strat- egy provides major challenges for policy and research. Several of these challenges are already taken up in the work programmes of the Seventh Framework programme and the Horizon 2020 research and innovation program of the EU. The definition of the term ‘bioeconomy’ is still a matter of discussion (see Schmidt et al., 2012), in part due to the fact that different countries have been using different defini- tions. For example, the OECD and the USA definitions are more focused on biotechnology than that of the EU. Different views are also linked to the use of the term bioeconomy in different branches of research, with different normative visions of technology. In particular, there is an emerging tension between an industrial/process vision of the bioeconomy and a territorial/landscape vision. In spite of this diversity, a common feature of the different visions of the bioeconomy is the central role of knowledge, research and innovation. From the point of view of tracing the impact of research, the bioeconomy goes in the direction of making the problem more “analytical” and more difficult. First, it tends to promote a stronger decomposition of material flows, and increasing the potential for re-combinations. In addition, it tends to promote cross-discipline and cross-sector “con- taminations”, making it more difficult to identify the flows of effects. Finally, working with biological materials tends to increase the degree of uncertainty about the characteristics of input and output, as well as of the outcome of research processes and their applications. A final and perhaps more important issue is that, given the knowledge-based nature of the bioeconomy, the impact of research critically depends on the stock of available knowledge and how it is mobilised to exploit newly produced knowledge. A symptom of the complexity and difficulty in providing clear-cut judgements is that the development of the bioeconomy is accompanied by a mix of opposite expectations about its effects and its role in ensuring sustainability. Pfau et al. (2014) identify different visions: (1) the assumption that sustainability is an inherent characteristic of the bioec- onomy; (2) the expectation of benefits under certain conditions; (3) tentative criticisms in light of potential pitfalls; and (4) the assumption that the bioeconomy impacts nega- tively on sustainability. This clearly hints at the potential opposite (ambiguous) effects of the bioeconomy on sustainability. A concept largely embedded in the bioeconomy discourse is that of bio-refinery. Biorefinery is the sustainable processing of biomass into a spectrum of bio-based products and bioenergy. It is based on the idea of a stepwise treatment of biomass starting from the 284 D. Viaggi highest added-value components down to energy (cascading approach) in such a way as to make the most out of limited biomass resource availability. Bio-refinery in Europe is promoted as one of the key economic organisational concept in the bioeconomy, in par- ticular for non-food chains. The research may focus on different issues in this field. One key focus is the extrac- tion of an increasing number of products from the same initial biomass; however, the qualifying feature of bio-refinery is the ordering of extraction, in such a way as to max- imise the value of production prior to the destruction of the biomass. This also goes in the direction of a higher level of disaggregation of biomass in more simple and basic components that can later be used as the building blocks of a variety of products. The degree of pureness of the obtained materials seems to have an added value here. At the same time, in many cases, the actual value of such products is uncertain due to the fact that they are “far” from final products, new to the market (or new substitutes of existing ones) or due to quality issues (presence of residues etc.). Moreover, technologies devel- oped by new research may yield a displacement of effects from one step to the other or change the structure of downstream processes after changing a step in the process. Final- ly, research can also lead to modified economic relationships, e.g. new products may have the potential to transform by-products into main products, justifying the process based on market forces, with implications in the allocation of effects. This effect is potential- ly more important as the number of potential extractions from the same raw material grows. As in all processes using biomass, which are often characterised by high transport costs per unit, spatial location is an essential feature of industrial organisation and the economic viability of processes. An emerging concept closely connected to the bioeconomy is that of the circu- lar economy, which relates to the idea that the economy should rely less on external raw materials and more on the re-use of resources that are already in the system. Chertow and Ehrenfeld (2012) define the circular economy and discuss the process of the organisation of circular economies in industrial symbioses. At present, the circularity of the economy (measured by the degree of circularity intended as the share of materials that flow back into the antropic system) is rather lim- ited. Hass et al. (2015) provide an estimation of the degree of circularity of the global and the EU-27 economy for 2005. They show that there is a global flow of roughly 4 giga- tonnes per year (Gt/yr) of recycled waste materials; this is of moderate size compared to 62 Gt/yr of processed materials and 41 Gt/yr of outputs. Biomass has an important role in this process as it accounts for 19 Gt/yr with only a 3% (7% in the EU-27) degree of circularity. One of the reasons for this is that biomass is largely used for energy purposes (including food), which makes it non-recyclable. Here, however, the definitions and the system considered are of significant impor- tance; if biomass is produced sustainably (that is, without damaging soil or water resourc- es and without depleting ecological carbon stocks), it can be considered renewable and the emitted CO2 as well as waste flows can largely be recycled into new primary biomass within ecological cycles (Jordan et al., 2007). Closing the cycle for biomass-related industries hence involves closing the cycle of nutrients (nitrogen and phosphorous), reducing waste from production and consumption, changing dietary patterns towards less meat demanding diets. As a result, both agriculture 285Research and innovation in agriculture: beyond productivity? and food are heavily affected by the process of getting to a more circular economy, or, looking at the other side of the coin, can directly contribute to it. The concept of the circular economy has implications not only in terms of the organi- sation of the economy, but also in relation to the role of technologies and, hence, research. Indeed, a growing focus of research is now related to closing the cycles of key materials, e.g. nutrients. This implies considering the degree of closeness of an economy as a key indi- cator of performance of the system or a change thereof as an important impact indicator for research. At the same time, the focus on circularity questions traditional productivity measures if they do not qualify the origin of raw materials (recycled or newly extracted). In economic terms, this implies investigating the different economic values attached to external versus re-used resources. Resources are re-used at a cost; external resources have a cost affected by extraction costs and scarcity rents. Both costs may change depend- ing on the quantity allocated to a given source or another. This hints at least at the idea that the degree of closeness of an economy may not tell the full story in economic terms, as actual values may be not proportional to the degree of closeness/openness. Instead, the value of resources in classical productivity measures can actually be more informative about the degree of optimal closeness of an economy, except in the case of externalities, market distortions or information failures. When market prices are not considered to be good estimates, shadow price approaches may help to investigate the actual degree of scar- city of resources (e.g. Halvorsen and Smith, 1984). Finally, an issue related to impacts in the circular economy and bioeconomy (i.e. using living organisms) is that scarce resources may in fact change over time, depending on the availability of resource stocks and technology. A pertinent example is the growing attention to the increasing scarcity of phosphorous in the field of agriculture. Hence, eventual indi- cators of circularity may require the flexibility to change over time to account for critical issues. Another issue that is typical of living organisms is their inherent variability, which should be taken into account in both evaluation and policy design (Carillo e Maietta, 2014). 2.2 Entrepreneurship and eco-innovation The role of the different actors in research and innovation is another issue that is changing dramatically. In a simplified (older) vision of research and technology uptake, the market and policies played major roles. The market was seen as a promoter of private research and the main determinant of technology uptake, whereas the public sector, for its part, was a research provider and promoter of targeted technology uptake, e.g. through subsidies. The transfer of innovation from science to the industry, initially thought of as a sort of linear process, has been over time supported by specific extension policies aimed at smoothing and encouraging this one-way flow. This naive idea of technology transfer has been widely challenged over time, nota- bly by investigating the number of actors, institutions and mechanisms that provide a bridge between research and the farming sector and considering feedback loops between the farming sector and research. The broad-based literature on Agricultural Knowledge and Innovation Systems (AKIS) indicates that this has become increasingly important in agriculture, especially in relation to the bioeconomy (Esposti, 2012). Relevant categories go beyond single extension units and farmers. Networks are becoming a more and more 286 D. Viaggi important category in this field. Networks also support a dynamic view of the innovation process, in which different types of actors can get involved at different stages and inno- vators need to continuously reflect and re-position themselves with respect to their envi- ronment. This view of the innovation process also implies that facilitators and innovation monitoring and evaluation methods can play a major role. In addition, it implies a re- thinking of policies to ensure that they are not excessively aimed at determining inno- vation direction, but rather support facilitation activities and the emergence of innova- tion initiatives (Klerkx et al., 2010). Accounting for higher and growing complexity also requires adequate approaches for research and evaluation. A pertinent example is that of the impact pathways evaluation approach (Douthwaite et al., 2003). Participatory approaches in different steps of the research to innovation processes have become particularly evident in the context of EU initiatives. The process started with the EU Technology platforms at the outset of the Seventh Framework Programme and has become paramount in the context of the new Horizon 2020 programme, focusing on the contribution to competitiveness through research and innovation and promoting a multi- actor approach in research projects. In this context, one of the most relevant phenomena in recent decades, which was also largely supported by the development of the bioeconomy, is the role of entrepreneur- ship in research and innovation. Entrepreneurship can take different forms, from broker- age of innovation as a specific business activity, to the financing of innovation, up to a sort of “innovation entrepreneurship”. Moreover, the process of developing ‘entrepreneurship’ activities and attitudes by researchers is increasingly being promoted. The role of entrepre- neurship has attracted attention in the field of biotechnology for at least two decades with the emergence of the term “bio-entrepreneurship” (Schoemaker and Schoemaker, 1998). The concept includes a wide range of typologies, ranging from researchers developing enterprises to exploit their knowledge to entrepreneurs investing in life science research and development companies. This concept is now widespread and increasingly used in specific educational programmes involving university students and researchers (see e.g. Uctu and Jafta, 2015). This pathway also draws attention to the process of building research objectives and priorities. A specific point concerns the growing role of businesses in building innova- tion strategies and, through these, guiding the development of oriented applied research. As this approach becomes more important, the building of circular connections between research and business seems to be giving additional weight to business, which is providing a vision for the future, while research has more and more the (limited) aim of buttressing this vision with much needed knowledge. From the point of view of impact evaluation, a key issue in recent years is the incor- poration of collective values (e.g. public goods concerns such as environmental and resource issues) into market strategies by the private sector. This can be viewed as the pro- gressive merging of private and public type values. This is witnessed by a process of strate- gic choice on the part of industry, awareness and related behavioural changes by consum- ers and the appropriate functioning of markets and marketing, including the communica- tion of values and the transmission of information about products and processes. In the incorporation of ‘green’ concerns into private business, a key concept is that of eco-efficiency. The concept attempts to reconcile economics and the environment. The 287Research and innovation in agriculture: beyond productivity? World Business Council for Sustainable Development (WBCSD) in 1992 defined eco-effi- ciency as being “ …achieved by the delivery of competitively-priced goods and services that satisfy human needs and bring quality of life, while progressively reducing ecological impacts and resource intensity throughout the life-cycle to a level at least in line with the earth’s estimated carrying capacity” (Schmidheiny, 1992, cited in Govindan et al., 2014). The green economy has been put forward as an even more positive and environmen- tally focused way of seeing the economy. It is intended as an economy seeking to reduce environmental and ecological impacts and that fosters sustainable development without degrading the environment. It also incorporates the idea of fairness. An interesting area of attention linking research and the green economy is provided by eco-innovation, which also offers examples of the articulated interplay between firm strategies, their economic context and inter-firm relationships. The exploration of the fac- tors of eco-innovation effort is considered to be at the heart of new research directions in the new millennium (Rashid et al., 2014), driven in particular by four eco-innovation drivers: regulatory push, technology push, market pull, and firm strategies. In the con- text of this trend, the connection between vision, research objectives and the impacts of technological innovation is becoming of central importance. For example, Björkdahl and Linder (2014) explain how and why a shared environmental vision can accelerate envi- ronmental innovation. Specifically, they emphasize that a shared environmental vision can lead to an increase in the number of application areas and in increased market sales based on existing green solutions. However, they also show that the efficacy of the shared vision is dependent on a good match between the environmental problems faced and the core competencies of the firm. Cuerva et al., 2014 found that the factors driving eco-innovation are different from other types of innovations. Based on a questionnaire carried out in the Spanish Food and Beverage sector, they found that technological capabilities such as R&D and human capi- tal foster conventional innovation, but not green innovation. On the contrary, the imple- mentation of Quality Management Systems (QMS) and differentiation contribute only to the adoption of green innovations. One of the findings of this study is that greater imple- mentation of voluntary certification schemes would be more effective in enhancing eco- innovation than public subsidies. Furthermore, attention to collaboration and the needs of consumers are positively associated with eco-innovation. Triguero et al. (2013), study- ing eco-innovation in Europe at the firm level, found that those entrepreneurs who give importance to collaboration with research institutes, agencies and universities, and to the increase in market demand for green products, are more active in all types of eco-inno- vation. Greater attention to existing regulations shape eco-product and eco-organisational innovations while expected regulations and access to subsidies and fiscal incentives do not have any significant effect on decisions to eco-innovate. Policy, management and communication instruments are playing a key role here as promoters of change. For example, quality management and certification schemes have been at the core of a wide field of research in recent decades and are increasingly the subject of analysis. New communication technologies may have a key role in boosting these connections, e.g. by improving the awareness of consumers and their ability to choose, or providing information, for example, about the quality and eco-friendliness of products. Examples include apps that help search for non-GMO, organic, or in-sea- 288 D. Viaggi son food1. The role of these technologies in shaping awareness, preferences and future demand-supply interaction is still largely unexplored. 2.3 Emerging methods for measuring the impacts of research: towards a diffuse life cycle assess- ment approach? Another strategy for understanding the implications of current trends in the analysis of the links between research and productivity is to look at instruments for measuring the impacts of (new) technologies and, indirectly, of research. In this perspective, a dominant role is currently played by Life-cycle assessment (LCA). LCA is an assessment method focusing on impacts generated by each unit of product (or, more appropriately, functional unit) along its lifecycle, from “cradle to grave”. The basis of the method is a compilation of the inventory of inputs and outputs, notably with reference to key resources (e.g. energy, water) or pollutants (e.g. GHG, nitrogen). Besides the inventory phase, LCA is being expanded to include the evaluation of dif- ferences among technology alternatives, including the use of multi-criteria analysis and the linkage with economic performance, e.g. using lifecycle costing, along the life cycle of a given product. LCA has been used for more than two decades as an environmental assessment tool and is increasingly used to support marketing messages. It is increasingly used as the basis for the selection of products in ‘green procurement’ and the inclusion of products in various national and regional eco-labelling schemes, i.e. European Flower, German Blue Angel, Nordic Swan eco-label etc. It is now widely promoted for early evaluation of research and innovation processes. In particular, LCA was already applied in the Sev- enth Framework programme of the EU (Tilche and Galatola, 2008) and is now regularly required in the calls of Horizon 2020. In principle, LCA responds to the basic idea of productivity, though expressed in a reverse way, i.e. aiming at minimising the unit of input and emission per unit of product. Notably, however, it tends to maintain a multidimensional and rather broad (and diverse) view of such a ratio, depending on the environmental indicators measured for input and output. From the point of view of the linkage between research and productivity, LCA addresses the need to better account for the broader impacts of technological research, hence considering a potentially wide range of effects on complex systems. In addition, it seeks to account for environmental effects along the value chain of a given product (bro- ken down in different key phases), hence being able to explicitly account for displacement or compensatory effects at different stages of the chain. It also takes into consideration by-products and recycling. Furthermore, LCA makes it possible to support prescriptions regarding the steps in the process where intervention/research is more urgent due to high- er criticalities in terms of impacts. For its characteristics, LCA appears especially suitable to address the impacts of inno- vation and research applied to the emerging issues discussed above (bioeconomy, circular 1 http://grist.org/list/this-powerful-app-brings-organic-farming-into-the-candy-crush-age/ http://foodtank.com/news/2013/10/twenty-three-mobile-apps-changing-the-food-system 289Research and innovation in agriculture: beyond productivity? economy and bio-refinery). In fact, several papers already report applications referred to these issues. For example, Mattila et al (2012) discuss the methodological aspects of apply- ing LCA to industrial symbioses and, more generally, to circular economies. The literature is also developing on the specificities of the application of LCA to bio-refinery systems (Sandin et al., 2015). LCA is also at the forefront of the measurement of eco-efficiency, which to date has been using different methods, ranging from simple indicators to modelling. LCA and DEA are arguably amongst the most used (Govindan et al., 2014; Lin et al., 2012). An example of the application in agriculture is available in Picazo-Tadeo et al. (2012). Yet LCA applications have also experienced difficulties and the above-referenced liter- ature emphasise a number of open issues. While databases are now more and more avail- able for standard applications of well-established technologies, coefficients needed for new technologies or for rapidly evolving product chains are often not readily available. Con- ceptual issues related to the boundaries of the system remain open, especially for not yet well-structured or evolving production chains. Moreover, LCA yields comparable results only when the functional unit is perfectly comparable, which weakens the potential for comparing different products or processes yielding different results in terms of services. Applications to the concepts of bioeconomy and biorefinery emphasise the issue of the attribution of impacts internally to the system considered; at the same time, as the process is purposely designed to yield multiple products, the issue of finding a common function- al unit or to allocate common effects to different products is also highlighted. Sandin et al. (2015) explore how the choice of the allocation methods influences results and in which decision contexts the choice is particularly important, by testing six allocation methods in a case study of a bio-refinery using pulpwood as feedstock. The results indicate that the choice of allocation method deserves careful attention, particularly in consequential stud- ies and in studies focussed on co-products representing relatively small flows. An issue of specific interest for this paper is that LCA tends to focus attention on environmental/resource use, whilst economic and social impacts remain more difficult to account for. Notably, recent attention has been devoted to the use of Life Cycle Costing (LCC), which is an economic assessment tool considering all projected significant and rel- evant cost flows over a period of analysis expressed in monetary value. It can be applied to a physical asset life or to the life cycle of a product/services in analogy to LCA. Nota- bly, this is gaining attention for use in public procurement, another area of innovation in which the purchasing power of public institutions is used to provide incentives towards environmentally friendly technologies (IISD, 2009; Dragos and Neamtu, 2013). From an economic point of view, two key nodes remain open issues, in particular in evaluating the impact of research. First, the way in which research can impact the produc- tion process includes several variables (e.g. uptake, organisation of the production process, concentration etc.), which makes it necessary to estimate potential impacts based on a number of assumptions. The use of LCA and location (including related limitations) must also be considered in the light of the different types of chains addressed: i.e. short, long, and global. Concepts connecting the measurement of impacts and trade relationships are also emerging, such as virtual water and water footprint. Second, the “engineeristic” impact in terms of changes in flows needs to be given an economic value. This also implies assumptions about, for example, the location of impacts. 290 D. Viaggi Taking the case of the use of water resources, abstraction can have unit costs of a different order of magnitude depending on the source used. It should also be acknowledged that the distribution of impacts across sectors can be non-neutral. An issue related to the economics of LCA information is its role in decision-making. On this issue, Sandin et al. (2014) note that, “particularly in inter-organisational R&D projects, the roles of LCAs tend to be unclear and arbitrary, and as a consequence, LCA work is not adequately designed for the needs of the project considered. There is a need for research on how to choose an appropriate role for LCA in such projects and how to plan LCA work accordingly” (Sandin et al., 2014, p. 97). Similarly, considering the con- nection with the work on the measurement of eco-efficiency, Govindan et al. (2014), draw attention to the need to better connect these studies to supply chain management. 2.4 From performance to positioning: sustainability and ecosystem services The wider context in which the above issues have been developed has been character- ised by the widespread use of the concept of sustainability as the increasingly important aim of agriculture and food systems and, related to that, of sustainability-oriented technol- ogy change. Without entering into the debate with respect to the definition of sustainability, it is clear from the literature that it is to a large extent related to socially constructed notions. On the one hand, the literature points out and advocates the need for the political pro- cess to define sustainability (Schepers, 2014). On the other hand, it is claimed that sus- tainability is a continuous social learning path and that such a transformation should be “profound (e.g. affecting moral standards and value systems), transversal (e.g. requiring the involvement of individuals as well as collective action) and counter-hegemonic (e.g. requiring the exposure and questioning of stubborn routines)” (Wals e Rodela, 2014). It is also important to emphasise that the problems encountered are not only related to providing adequate definitions, but also touch upon the empirical conceptualisation and practical measurement of sustainability, including its relationship with globalisation and development literature (Olson et al., 2014). A good example of the practical problems faced in the measurement of sustainability is given by the approaches towards the assessment of environmental sustainability of agri- culture, somehow the component of sustainability better studied, as compared to economic and social sustainability. This issue has been largely addressed in the literature through the use of indicators (Bockstaller et al., 2008). As the number of indicators developed is both high and varied, the literature has also formulated a number of proposals for composite indicators and integrated sustainability assessments (Rodrigues et al., 2010a; 2010b). More- over, a number of methodologies have been developed relying not only on more or less complex (sets of ) indicators (Singh et al., 2009), but also on modelling (Ness et al., 2007). The literature notes the contradiction among the different requirements expressed towards these tools, which are expected to be at the same time specific yet broad, tailored but standardised (Ness et al., 2007). There is also demand for composite indexes, whilst recognising that simple aggregate indexes can provide misleading information to decision- makers (Singh et al., 2009). In most cases, data availability remains the clearest criterion for decisions about the individual indicators and tools to be used. Bockstaller et al. (2008) 291Research and innovation in agriculture: beyond productivity? conclude that, as the data available at the regional level are usually limited, several sim- ple indicators should be used, at least at this level. Only when more detailed information is available can indicators based on operational models be useful. In experimental studies, when possible, it is suggested to use both measured indicators and model-based indicators. In recent years, concepts such as resilience and vulnerability have increasingly accom- panied or replaced that of sustainability. The detailed consideration of each of these terms would require the examination of a wide body of literature, which is beyond the scope of this paper. Notably, all of these issues have found noteworthy parallel use in ecology, environmental economics and development economics, highlighting the importance of dynamics and the relevance of “potential” effects/changes. Another feature of these con- cepts is their attempt to be comprehensive, which is at times pursued at the cost of diffi- culties with accurate definitions and measurement. The most direct implication for productivity measures arising from this field of study is a push for a stronger consideration of sustainability in both output and resource con- cerns. It should, however, be emphasised that expectations regarding easy corrections of traditional productivity measures may be overambitious. One of the consequences of the increasing uncertainty in the different interpretations of ‘sustainability’ and the difficul- ty in the measurement of outcomes/impacts of processes, accompanied by the quest for more comprehensive concepts, is the move towards proxies able to measure the position- ing towards the future, rather than the direct measurement of outcomes/impacts either ex- ante or ex-post. As a consequence, also in relation to the measurement of productivity and of the effects of research, the consideration of proxies, or of a limited number of selected issues (most likely measuring pressures), could remain the more realistic option. A concept closely related to sustainability assessment is that of Ecosystem Services (ES). According to TEEB (2010), ES are the direct and indirect contributions of ecosys- tems to human well-being. They are most often categorized into four types (MEA, 2003; TEEB, 2010): provisioning, regulating, habitat and cultural services. In contrast to other approaches, the ES approach takes ecosystems directly into account and links them to the uses that human beings can make of the services they provide. The approach has gained a broad consensus and has increasingly been adopted in policymaking. The use of ES pre- sents several advantages, among which, notably for the scope of this paper, to include in the same framework services directly linked to “traditional” productivity measures (provi- sioning) and those that relate to other ecosystem roles in human life, thus making explicit the various trade-offs, synergies and relative weights. On the other hand, ES make it pos- sible to link issues related to the economics of sustainability with the ecosystem context and to cast sustainability issues in a territorial and landscape dimension (van Zanten et al., 2014). Of significant importance is also the fact that the concept seems to be particu- larly suitable for policy communication. On the other hand, a number of studies show that the understanding of ES, and their inherent trade-offs, require an enlargement of the system considered and hence re-intro- duces, and even emphasises, the trade-offs between detail and comprehensiveness. Fur- thermore, ES does not solve data issues, which remain a key driver of the choice of spe- cific indicators and their measurement strategies, nor do they address the problem related to the attribution of economic values to ES services, at least for the part represented by public goods-type services and externalities (Viaggi, 2015). 292 D. Viaggi Another issue linked to sustainability is related to the strategies currently being used to achieve improved product sustainability. Far from traditional end-of-pipe approaches, current strategies directly address process and product design, value chain organisation or even producer habits. This is the focus of a wide body of literature on alternative technol- ogy options in agriculture, but it is now also of increasing importance for food produc- tion. In this regard, van der Goot et al. (2016) list several technological strategies, includ- ing: avoiding dilution, minimizing drying, focusing ingredient production on functional- ity rather than purity, tailoring ingredient production to specific applications rather than for general use, developing smaller and more flexible fractionation processes which should also be located in the vicinity of the application, and using milder process conditions for less refined ingredients. In other words, sustainability concerns are increasingly embedded in the whole product design and chain/system functioning. Similar approaches are advocated for most of the issues connected to sustainabil- ity, including the promotion of the circularity of the economy. These considerations have implications in particular for the measurement of impact in research aimed at improving sustainability, which in fact tends to require an analysis (and awareness) of entire process- es and a comprehensive view of the related economic dimensions. 3. Discussion, implications for research and the way forward Each of the issues and perspectives illustrated above has potential specific implica- tions for the relationship between research, innovation and productivity and for the meas- urement/evaluation of the impacts of research. The implications can be broadly organised into three interconnected topics: a) measurement of productivity; b) parameters able to explain changes in productivity, especially due to research; c) ways of making the connection between research efforts and productivity change. As far as the measurement of productivity is concerned (point a) the emergence of environmental and resource problems, and the new areas of concern illustrated above, have primarily brought to the attention (even more strongly than before) the need to expand the range of output indicators and the range of resources to be con- sidered as input. Besides this, the current trends seem to introduce novelties in the quality of the effects sought, which are increasingly represented by: a) soft effects, amenable to interpretations and flexibility in the social construction of related values (and hence prices), and; b) effects that are more valued for their potential than for their actual observable effects. Altogether, while the classical measures of productivity remain key references at the aggregate level, the emerging attention to circular economies and resource efficien- cy brings into question, in particular, the application at intermediate and lower levels of aggregation (chains, firms, small territorial units). Here, the widespread diffusion of LCA signals (and partially responds to) the need to develop more functional measures of the effects of (and in) different steps in the chain, and brings directly to attention the tech- nological and economic inter-linkages among different processes. LCA is itself, however, challenged by the emphasis on circularity and by the social construction of values related to impacts. Indeed, the discrepancy between more accurate technical measures of impact 293Research and innovation in agriculture: beyond productivity? (or at least pressures) and the limited ability to valuate their effects in economic terms is still very high. It is likely that the new issues emerging from the development of the bio- economy will further contribute to this discrepancy, making the effects of research even more difficult to measure objectively. These reflections also lead to question the extent to which economics can actually play a role in measuring productivity effects in light of current technological development. On the one hand, many issues addressed above, such as circularity in the economy, can already be incorporated into the measures currently being used, if prices work well, particularly with respect to scarcity. On the other hand, it is becoming more difficult to account mon- etarily for the preferences of final consumers, who are becoming less stable and more driv- en by perceptions, expectations and information distortions (especially for services that are less and less related to basic needs or for those that are new for consumers). It is also becoming difficult to identify the economic mechanisms that are determining the impact of research, starting with the use of new knowledge and the adoption of new technologies. In this context, the traditionally clear distinction between private and public goods and the category of externalities (and the related instruments for economic valuation), are also weakening due to the “marketization” of environmental and social values, the increas- ingly explicit socially (or policy) driven construction of values and preferences and the growing number of cases of goods that are somewhere between private and public. These trends expose the system to the instability of preferences over time, which may lead to dif- ficulties in prediction, overlapping and double counting. They also require (and partially ensure) greater attention to information and communication, the embedding of technol- ogy change in participatory processes, and the awareness of the processes leading to the construction of values. Similarly to output measurement, the interpretation and understanding of determi- nants of productivity (point b), and among them research, also require effort to account for an increasing number of parameters that contribute to the explanation of productiv- ity changes. This includes accounting better for a larger number and diversified quality of input, more complex interplays among sectors and larger geographical interactions. A specific element requiring attention is the strengthening of the connection between research and innovation efforts and the stronger linkages with human capital and entre- preneurship attitudes. Changes in the way research is performed must also be better analysed. Research is becoming increasingly analytical, cross sector and taking on multiple methods of inter- action (in interdisciplinary, multidisciplinary and transdisciplinary contexts). It is also necessary to take into account the multiple number of actors, which also imply different ways of understanding the directions of interaction among different pieces of knowledge. In this framework, information about research expenditure, even when available, becomes less and less satisfactory in accounting for research efforts, without considering the increasing difficulties in linking expenditures to the specific objectives of research (and hence expectations about the direction of its impacts). Furthermore, specific attention should be paid to understanding how to account for the existing stock of knowledge. Past research expenditure is less and less a good way for accounting for the increased stock of very diversified knowledge; this issue in itself deserves to be the subject of more focused research, considering improved proxies and a 294 D. Viaggi better understanding of mechanisms taking into account spatial dynamics, including flows of knowledge and technology. The connection between research and its effects (point c), is possibly the most puz- zling of the three topics listed above, and its evolution is still less studied compared with performance measurement and its potential determinants. In addition, recent literature shows that addressing this link is not only a matter of measuring items, but more widely of understanding and codifying cause-effect relationships among factors affecting the direct and feed-back loops between research, technology adoption and performance change. Relevant issues to be accounted for include scope and scale of the evaluation, i.e. what is the right bundle of products/issues to be addressed and at what geographical level and process detail. Bioeconomy research is a good example of the need to better understand how to address these issues: innovation tends to be increasingly analytical in improving processes as its effects are becoming “finer” and there is a tendency toward a greater dis- aggregation of biomass into smaller blocks. But some concerns do exist (among which sustainability, counteracting this trend). The joint pathway of the circular economy and biorefinery is leading towards the need for more analytical and more comprehensive approaches. Yet their effects and, arguably, the perception of their effects has become more ambiguous and difficult to measure. A more specific issue relates to time lags for the effects of research. Most of the lit- erature emphasises that long time lags are necessary to allow research to impact on pro- ductivity (Alston, 2010; Wang, 2013). However, the evolution of the bioeconomy and the explicit focus on innovation in research programmes (see Horizon, 2020) may be expected to push for reducing such lags, or, more likely, to diversify even more the lags among the various pieces of research. This variety of potential time lags is also visible from having a simultaneous quest for very applied research supporting short-term innovation and a very open “blue sky” (in the words of the EU Commission) and fundamental research. In addi- tion, given the status of continuous change and the different technologies that interact, the role of both indirect and unexpected effects, which may in fact occur at different points in time, are more prominent. Another economic issue, stemming in part from the evolution in governance/legisla- tion and the existence of more sophisticated technologies, is the growing complexity of property rights in research results. This makes the use of research results and the appro- priation of their benefits more distributed, while the participation in the exploitation and the potential for alternative uses of knowledge also become wider whilst, at the same time, requiring more flexible ways of collaboration. Given this complexity of chain effects, one of the strategies for future understand- ing of the impact of research would be to look at shorter chains of effects, i.e. between research and adoption/organisation, or between adoption and performance, rather than trying to address the long way between research and final changes in the system, with the related causality problems. Altogether, one could argue that, while the productivity of research should remain and even become a more important focus of research, the real priority issue is not the measurement of impacts (though it is to some extent), but rather the mechanisms making the link between research and its impacts, which could also be considered as a good (or the best) proxy of impact itself, or at least as a measure of potential impact. 295Research and innovation in agriculture: beyond productivity? Several pathways for further research crosscutting the simplified three points illustrat- ed above can be identified. These include, notably, the following: 1. There is a need to identify improved representations of goods, technologies and their change that can be used to better account for the current trends in technologies that break down biomass and recombine its compounds; one direction could be in repre- senting input as bundles of attributes (elements) and the output as potential attributes rather than products. 2. There is also a need to better investigate new institutional mechanisms, especially the role of entrepreneurship in research, knowledge exploitation and the social construc- tion of successful technologies. This connects with the notion of business models and the “shape” (or, better, “non-shape”) of enterprises (which, metaphorically, can be bet- ter represented by an amoeba rather than any other constant form concept) becoming more and more a bundle of loosely connected rights and values; this trend is empha- sised by the growing complexity in technology ownership and the exploitation of innovation. 3. A better study of the interplay between consumers and producers, beyond simplified market mechanisms, is also needed. On this issue, new awareness building and mar- keting strategies (social networks) and thoroughly informed/aware communities (yet often limited in their intepretation ability) connecting demand and supply, boosted by new technologies (e.g. smartphones) call for the need for collaboration and link- ages between studies on production and studies on consumption in order to directly deal with uncertainty about prices and market shares. This is particularly relevant for new technologies for which the market has yet to be developed, in which frequently changing or new policy measures are also often observed. 4. New communication technologies have a key role in several of the processes men- tioned above; though they are playing a disruptive role in changing societies, their role as means and promoters of change in economic behaviour linked to innovation and exploitation of research is still insufficiently studied. 5. There is also a need for studies on tools and strategies to deal with uncertain futures. A specific issue related to emerging technologies and their interplay with awareness and market building is the difficulty in predicting futures; trends are less and less a relevant indication for the future, while a breakthrough of new solutions may be more relevant. On this point, economic valuations using potential trend-breaking scenarios may require greater attention with respect to on past-based expectations. 6. The investigation of new potential tools to measure productivity and link with research is also needed. The existing tools are for the most part well-established in the literature; sometimes they are seen to be novel when “discovered” by different dis- ciplinary fields or when new variants become available, but in the last couple of dec- ades there have mostly been incremental innovations rather than radical innovations in this field. Contamination among existing tools could be a promising pathway and is already under way, e.g. studies proposing combinations of LCA and DEA analysis, or LCA and Multi-criteria analysis. 7. This goes hand by hand with the further development of studies to better under- stand research and technology impact pathways in this changing context; a distin- guishing feature of future studies could be found in a more extended use of impact 296 D. Viaggi pathways approaches, to better account for the larger set of determinants discussed in this paper. 8. An investigation into the new role of policy and related mechanisms is also needed. Not only regulation and policy-defined prices contribute to reveal and signal prefer- ences through the value chain; the building of new markets is becoming a stronger policy issue that has a broad range of implications. Indirect policy mechanisms, such as those helping to reveal preferences related to public goods or facilitating coordi- nation, are becoming more important, yet continue to be poorly understood. An increased emphasis on policy measures related to awareness, information, education and knowledge management may also result in the need for different tools to study policies, e.g. more qualitative and systemic tools. 9. Information and data needs remain a key issue for researchers in the field of the evaluation of research impacts on productivity. Gaps primarily involve productivity measures, in particular “non-conventional” components of productivity, such as envi- ronmental improvements considered in multi-output productivity specifications or input-saving approaches. Data gaps are also very much relevant for research expendi- ture and potential intermediate explanatory variables (e.g. AKIS actors). However, it should be emphasised that this issue cannot be restricted to complaints regarding insufficient data availability. Rather, data users and providers should collaborate in co- construct data sources taking into account forward looking needs. In addition, greater attention should be given to emerging sources of data due to the digitalisation of huge amounts of information and related opportunities for data mining and linkages of databases (still largely unexploited). 4. Conclusions This paper has investigated the implications for research, innovation and productivity studies of selected bioeconomy-related issues, namely: a) the concepts of bioeconomy, cir- cular economy, resource efficiency and bio-refinery; b) the connection with entrepreneur- ship and eco-innovation; c) changing tools in research assessment, in particular the wide- spread use of LCA; and d) the evolving concepts of sustainability and ecosystem services. We argue that the “traditional” idea of productivity intended as an output/input ratio maintains (and may strengthen) its role on the aggregate, though the current trends in research are more focused than they were in the past on creating “potential” rather than straightforward changes. Furthermore, we find that, while the role of research in produc- tivity change is likely increasing, it is becoming somewhat more difficult to link changes in productivity to specific research actions. As a result of the above, the understanding of the interconnections and pathways between research and productivity are becoming more relevant than final productivity measures themselves, though much more difficult to trace than in the past. Policy and practitioners are requiring improved ways of performing ex-ante and ex-post analyses that are better connected with decision making and capable of managing interplays between aggregate and disaggregate levels. This is also in line with a growing emphasis on incorporating sustainability concerns into private (both firm and consumer) behaviour while dealing with globalised markets and global environmental and social concerns. 297Research and innovation in agriculture: beyond productivity? This highlights a number of new challenges for research in economics, especially related to agriculture, food and the bioeconomy. A key issue concerns the need for meth- odological developments, which may find their basis in the improved knowledge of new research processes and new technologies, as well as in a better understanding of surround- ing societal change. It also requires economists to take up the “procedural” and cultural challenges of an increased involvement in the agriculture and bioeconomy innovation system, while at the same time guaranteeing objective analyses and robust independent judgements, suitable for evidence-based support to decision-making. Acknowledgments The author wishes to thank Ornella Maietta and Laurens Klerkx for their useful com- ments and suggestions on intermediate versions of this paper. An earlier version of this paper was presented at the 4th AIEAA Conference, Ancona, 11-12 June 2015. The usual disclaimers apply. References Alston, J., Norton, G.W. and Pardey, P.G. (1995). Science under scarcity. Principles and practice for agricultural research evaluation and priority setting. CAB international. Alston, J. M., Andersen, M. A., James, J. S. and Pardey, P. G. (2011). The Economic Returns to U.S. Public Agricultural Research. American Journal of Agricultural Eco- nomics 93 (5): 1257-1277. doi:10.1093/ajae/aar044. Alston, J.M., Andersen, M.A., James, J.S. and Pardey, P.G. (2010). Persistence Pays: U.S. Agricultural Productivity Growth and the Benefits from Public R&D Spend- ing. Persistence Pays: U.S. Agricultural Productivity Growth and the Benefits from Public R&D Spending. http://www.scopus.com/inward/record.url?eid=2-s2.0- 74549206738&partnerID=tZOtx3y1. Björkdahl, J. and Linder, M. (2014). Formulating Problems for Commercializing New Technologies: The Case of Environmental Innovation. Scandinavian Journal of Man- agement 31 (1): 14-24. doi: 10.1016/j.scaman.2014.05.001. Bockstaller, C., Guichard, L., Makowski, D., Aveline, A., Girardin, P., and Plantureux, S. (2008). Agri-Environmental Indicators to Assess Cropping and Farming Systems. A Review. Agronomy for Sustainable Development 28 (1): 139-149. doi: 10.1051/ agro:2007052. Carillo, F. and Maietta, O.W. (2014). The relationship between economic growth and envi- ronmental quality: the contributions of economic structure and agricultural policies. New Medit 13 (1): 15-21. Chen, S. and Golley, J. (2014). ‘Green’ Productivity Growth in China’s Industrial Economy. Energy Economics 44 (July): 89-98. doi: 10.1016/j.eneco.2014.04.002. Chertow, M. and Ehrenfeld, J. (2012). “Organizing Self-Organizing Systems.” Journal of Industrial Ecology 16 (1): 13-27. doi: 10.1111/j.1530-9290.2011.00450.x. Coelli, T., Rao, D., O‟Donnell, C. and Battese, G. (2005). An Introduction to Efficiency and Productivity Analysis. Springer, New York, second edition. 298 D. Viaggi Cuerva, M. C., Triguero-Cano, A. and Córcoles, D. (2014). Drivers of Green and Non- Green Innovation: Empirical Evidence in  Low-Tech SMEs. Journal of Cleaner Pro- duction 68 (April): 104-113. doi: 10.1016/j.jclepro.2013.10.049. De Wilt, J. G., Diederen, P. J. M., Butter, M. and Tukker, A. (2001). Innovation Challenges for European Agriculture. Foresight 3 (4): 341-352. http://www.scopus.com/inward/ record.url?eid=2-s2.0-3543122960&partnerID=tZOtx3y1. Douthwaite, B., Kuby, T., van de Fliert, E. and Schulz, S. (2003). Impact Pathway Evalua- tion: An Approach for Achieving and Attributing Impact in Complex Systems. Agri- cultural Systems 78 (2): 243-265. doi: 10.1016/S0308-521X(03)00128-8. Dragos, D. and Neamtu, B. (2013). Sustainable Public Procurement: Life Cycle Costing (LCC) in the New EU Directive Proposal. EPPPL, 1: 19-30. Esposti, R. (2012). Knowledge, Technology and Innovations for a Bio-based Economy: Lessons from the Past, Challenges for the Future. Bio-based and applied economics 1 (3): 235-268. European Commission (2012a). Innovating for Sustainable Growth: A Bioeconomy for Europe, SWD (2012). 11 final. Brussels, 13.2.2012. COM(2012) 60 final. European Commission (2012b). Commission staff working document accompanying the “Communication on Innovating for Sustainable Growth: a Bioeconomy for Europe”, Brussels. Fuglie, K. (2015). Accounting for growth in global agriculture, Bio-based and applied eco- nomics 4(3): 201-234. Gaunand, A., Hocdé, A., Lemarié, S., Matt, M. and de Turckheim, E. (2015). How Does Public Agricultural Research Impact Society? A Characterization of Various Pat- terns. Research Policy 44 (4): 849-861. doi: 10.1016/j.respol.2015.01.009. Govindan, K., Sarkis, J., Chiappetta Jabbour, C.J., Zhu, Q. and Geng, Y. (2014). Eco-Effi- ciency Based Green Supply Chain Management: Current Status and Opportuni- ties. European Journal of Operational Research 233 (2): 293-298. doi: 10.1016/j. ejor.2013.10.058. Haas, W., Krausmann F., Wiedenhofer, D. and Heinz, M. (2015). How Circular Is the Global Economy? An Assessment of Material Flows, Waste Production, and Recy- cling in the European Union and the World in 2005. Journal of Industrial Ecology 19 (5): 765-777. doi: 10.1111/jiec.12244. Halvorsen, R. and Smith, T.R. (1984). On measuring natural resource scarcity, Journal of Political Economy 92 (5): 954-964. International Institute for Sustainable Development (IISD) (2009). Life Cycle Costing in Sustainable Public Procurement: A Question of Value. IISD, Winnipeg, Manitoba, Canada, 28 pp. Klerkx, L., Aarts, N. and Leeuwis, C. (2010). Adaptive Management in Agricultural Inno- vation Systems: The Interactions between Innovation Networks and Their Environ- ment. Agricultural Systems 103 (6): 390-400. doi: 10.1016/j.agsy.2010.03.012. Latruffe, L. (2010). Competitiveness, Productivity and Efficiency in the Agricultural and Agri-Food Sectors, OECD Food, Agriculture and Fisheries Papers No. 30. Maietta, O. W. (2015). Determinants of university-firm R&D collaboration and its impact on innovation: A perspective from a low-tech industry. Research Policy 44 (7): 1341- 1359. 299Research and innovation in agriculture: beyond productivity? Mattila, T., Lehtoranta, S., Sokka, L., Melanen, M. and Nissinen, A. (2012). Methodologi- cal Aspects of Applying Life Cycle Assessment to Industrial Symbioses. Journal of Industrial Ecology 16 (1): 51-60. doi: 10.1111/j.1530-9290.2011.00443.x. MEA (2003). Ecosystems and Human Well-being. A Framework for Assessment. Ness, B., Urbel-Piirsalu, E., Anderberg, S. and Olsson, L. (2007). Categorising Tools for Sustainability Assessment. Ecological Economics 60 (3): 498-508. doi: 10.1016/j. ecolecon.2006.07.023. Olsson, L., Hourcade, J.-C. and Kohler, J. (2014). Sustainable Development in a Glo- balized World. The Journal of Environment & Development 23 (1): 3-14. doi: 10.1177/1070496514521418. Pfau, S., Hagens, J., Dankbaar, B. and Smits A. (2014). Visions of Sustainability in Bioec- onomy Research. Sustainability 6 (3): 1222-1249. doi: 10.3390/su6031222. Picazo-Tadeo, A.J., Beltrán-Esteve, M. and Gómez-Limón, J.A. (2012). Assessing Eco- Efficiency with Directional Distance Functions. European Journal of Operational Research 220 (3): 798-809. doi: 10.1016/j.ejor.2012.02.025. Rashid, N., Jabar, J., Yahya, S. and Shami, S. (2014). Dynamic Eco Innovation Practices: A Systematic Review of State of the Art and Future Direction for Eco Innovation Study. Asian Social Science 11 (1): 8-21. doi: 10.5539/ass.v11n1p8. Rodrigues, G.S., Rodrigues, I.A., de Almeida Buschinelli, C.C. and de Barros, I. (2010a). Integrated Farm Sustainability Assessment for the Environmental Management of Rural Activities. Environmental Impact Assessment Review 30 (4): 229-239. doi: 10.1016/j.eiar.2009.10.002. Rodrigues, G.S., Rodrigues, I.A., de Almeida Buschinelli, C.C. and de Barros, I. (2010b). Integrated Farm Sustainability Assessment for the Environmental Management of Rural Activities. Environmental Impact Assessment Review 30 (4): 229-239. doi: 10.1016/j.eiar.2009.10.002. Sandin, G., Røyne, F., Berlin, J., Peters, G.M. and Svanström, M. (2015). Allocation in LCAs of Biorefinery Products: Implications for Results and Decision-Making. Jour- nal of Cleaner Production 93 (April): 213-221. doi: 10.1016/j.jclepro.2015.01.013. Sandin, G., Clancy, G., Heimersson, S., Peters, G.M., Svanström, M. and ten Hoeve, M. (2014). Making the Most of LCA in Technical Inter-Organisational R&D Projects. Journal of Cleaner Production 70 (May): 97-104. doi: 10.1016/j.jclepro.2014.01.094. Schepers, S. (2015). Managing the Politics of Innovation and Sustainability. Journal of Pub- lic Affairs 15: 91-100. doi: 10.1002/pa.1521. Schmidheiny, S. (1992). Changing course: A global business perspective on development and the environment, The MIT press, Cambridge MA, pp. 448. ISBN: 9780262193184. Schoemaker, H. J. and Schoemaker, A. F. (1998). The three pillars of bioentrepreneurship. Nature Biotechnology 16 (Suppl): 13-15. Retrieved from http://www.scopus.com/ inward/record.url?eid=2-s2.0-0032067967&partnerID=tZOtx3y1 Singh, R.K., Murty, H.R., Gupta, S.K. and Dikshit, A.K. (2009). An Overview of Sustaina- bility Assessment Methodologies. Ecological Indicators 9 (2): 189-212. doi: 10.1016/j. ecolind.2008.05.011. Spiertz, H. (2014). Agricultural Sciences in Transition from 1800 to 2020: Exploring Knowledge and Creating Impact. European Journal of Agronomy 59 (September): 96-106. doi: 10.1016/j.eja.2014.06.001. 300 D. Viaggi Swinnen, J. and Weersink, A. (2013). Challenges and policy options in the global bio- economy: Introduction and overview. Agricultural Economics 44: 379-380. TEEB (2010). The economics of ecosystems and biodiversity: mainstreaming the economics of nature: a synthesis of the approach, conclusions and recommendations of TEEB. Tilche, A. and Galatola, M. (2008). Corner ‘EU Life Cycle Policy and Support.’ The Interna- tional Journal of Life Cycle Assessment 13 (2): 166-167. doi: 10.1065/lca2008.02.378. Triguero, A., Moreno-Mondéjar, L. and Davia, M.A. (2013). Drivers of Different Types of Eco-Innovation in European SMEs. Ecological Economics 92 (August): 25-33. doi: 10.1016/j.ecolecon.2013.04.009. Uctu, R. and Jafta, R. C. C. (2013). Bio-entrepreneurship as a bridge between science and business in a regional cluster: South Africa’s first attempts. Science and Public Policy 41 (2): 219-233. doi: 10.1093/scipol/sct049. van der Goot, A.J., Pelgrom, P.J.M., Berghout, J.A.M., Geerts, M.E.J., Jankowiak, L., Hardt, N.A., Keijer, J., Schutyser, M.A.I., Nikiforidis, C.V. and Boom, R.M. (2016). Con- cepts for Further Sustainable Production of Foods. Journal of Food Engineering 168 (January): 42-51. doi: 10.1016/j.jfoodeng.2015.07.010. van Zanten, B.T., Verburg, .H., Espinosa, M., Gomez-y-Paloma, S., Galimberti, G., Kan- telhardt, J., Kapfer M., Lefebvre M., Manrique, R., Piorr, A., Raggi, M., Schaller, L., Targetti, S., Zasada, I., Viaggi, D. (2014). European Agricultural Landscapes, Com- mon Agricultural Policy and Ecosystem Services: A Review. Agronomy for Sustain- able Development 34 (2): 309-325. doi: 10.1007/s13593-013-0183-4. Viaggi, D., Mantino, F., Mazzocchi, M., Moro, D. and Stefani, G. (2012). From Agricultur- al to Bio-based economics? Context, state-of-the-art and challenges. Bio-based and Applied Economics 1: 3-11. Viaggi, D. (2015). Editorial. Journal of Environmental Planning and Management 58 (12): 2082-2087. doi: 10.1080/09640568.2015.1084832. Wang, S.L., Heisey, P.W., Huffman, W.E. and Fuglie, K.O. (2013). Public R&D, Private R&D, and U.S. Agricultural Productivity Growth: Dynamic and Long-Run Relation- ships. American Journal of Agricultural Economics 95 (5): 1287-1293. doi: 10.1093/ ajae/aat032. Yin, K., Wang, R., Zhou, C.. and Liang, J. (2012). Review of Eco-Efficiency Accounting Method and Its Applications. Acta Ecologica Sinica 32 (11): 3595-3605. doi: 10.5846/ stxb201104280564. Zilberman, D., Kim, E., Kirschner, S., Kaplan, S. and Reeves, J. (2013). Technology and the Future Bioeconomy. Agricultural Economics 44 (s1): 95-102. doi: 10.1111/agec.12054.