The Rationale for Defra Investment in R&D Underpinning the Genetic Improvement of Crops
The rationale for Defra investment in R&D underpinning the genetic improvement of crops and animals(IF0101)
FINAL report TO DEFRA
Dominic Moran§
Andrew bARNES
Alistair mcvittie
SAC COMMERCIAL LTD
SAC LAND ECONOMY AND ENVIRONMENt GROUP
12 March 2007
§
1
Acknowledgements
Many people have given their time and forbearance in the preparation of this report. We would particularly like to thank participants in our original discussion groups, and those who took the trouble to respond to our electronic survey of the wider research community and stakeholders. Geoff Simm and Roger Sylvester Bradley were instrumental in marshalling the outputs of the expert discussion groups. Peter Amer (AbacusBio Ltd) provided the basic model onto which we have pasted further assumptions that underlie the livestock analysis. Steve Hoad fielded numerous questions about plant science with good grace. The views and conclusions expressed necessarily echo the views we have heard during this process. But their expression here is entirely our own.
Executive Summary
Defra is facing an expanding range of environmental challenges that may require reorientation of its policy and spending priorities. With this shift in emphasis comes a need to reconsider policy options for meeting these objectives and, by extension, the weight put on different forms of research and development (R&D) that provide a policy evidence base.
This policy shift inevitably poses questions about the government role in addressing market failure, and delineating responsibility for delivering public versus private good outcomes. This in turn leads to questions about the role and effectiveness of research, in this case in genetics, as a proximate investment option for delivering public good objectives. An important consideration is whether a genetic approach can respond to government priorities, and whether it can be judged as economically efficient.
Clarifying this question requires an exercise in research prioritisation; a process of clarifying policy objectives, specifying and prioritising options to meet these objectives, and appraising their economic viability. This study sets out the process of evaluating the role of plant and animal genetic research against new policy objectives. It then attempts to provide a forward-looking evidence-base on the economic efficiency of prospective Defra-funded R&D options in plant and animal genetics.
There are strategic reasons why government might want to maintain a presence in genetic research in plants and animals, but the value of the options afforded by a genetic research base are not easy to quantify. Most of the existing studies on economic returns to agricultural R&D are ex-post - i.e. looking backwards. While these are informative, they tend to be based on conventional private productivity (i.e. yield) returns, which are not the focus of Defra's new predominantly public goods priorities. Much less literature is focussed on genetic traits that confer public good benefits, or is of an ex-ante variety. Some studies of the potential returns to transgenic crops can provide indicative information on yield returns.
Ex-ante analysis has its own drawbacks in the extent of subjectivity attached to the methodological approach to selecting priorities, and predicting future research outcomes. It can also be further complicated by the need to place a value on the public good outputs of research, which are predominantly non-market welfare effects.
Methodological literature on the evaluation of R&D points to the dangers of cherry-picking research results to evaluate. It is also cautious about the extent to which a rate of return approach can provide sufficient policy evidence. This is because it can be artificial to assess the returns to a project in isolation from the wider research environment in which it has been developed, conducted, and ultimately used. Indeed, intended and ultimate use made of genetic research can sometimes diverge significantly. Much basic and applied science research often gives rise to wider social benefits than are not captured in an obvious observable outcome. To this extent, the linear approach adopted here probably underestimates the returns to public investment.
Despite the difficulties observed in rate of return analysis, it is important to attempt to demonstrate returns, and especially in the case of non market returns that are becoming more central to policy.
This study combined an options appraisal and an economic appraisal of candidate research to target Defra policy objectives. Rather than a detailed inventory of research options, the aim was to give an overall sense of the economic efficiency of genetic based R&D. The options appraisal considered and scored a long list of research ideas and themes that could be matched with a specific policy objective (e.g. climate change mitigation, water quality, biodiversity etc). The resulting shortlist was then subjected to a more detailed economic appraisal that consisted of a cost-benefit and a cost-effectiveness analysis of selected research areas. Both stages were informed by expert and stakeholder input.
The two basic questions are as follows. First, for the candidate research, and making a distinction between private and public good returns, is there evidence that benefits or public good returns to research outweigh the costs to the extent normally required for public sector spending decisions? The second question concerns the relative cost-effectiveness of the candidate research. Specifically, for the attainment of a given amount of a given objective, is a genetics approach the least cost means of delivery?
The first question is a more exacting approach to appraising public spending on research. But the indicator commonly derived (economic rate of return) by this method, does not necessarily indicate whether interventions yield near term benefits or the relative magnitude of the returns. Accordingly, the study undertakes a comparison of relative costs of different methods for achieving specified objective outcomes, with a preference for larger outcome delivery in the near term. These alternative methods could include other research avenues, or other policy levers such as market-based instruments, all of which have different costs implications.
On the cost-benefit comparison, our analysis suggests that both animal and plant genetic research targeting public good outcomes, can yield rates of return that stand comparison with other public sector investments. While the distinction between public and private returns is arguable, under reasonable assumptions, animal genetic improvement can yield rates of return ranging between 11 - 18%. Plant research may deliver rates ranging 21-61%. These rates of returns are calculated over a limited time horizon and compare to a standard Treasury discount rate of 3.5%. They compare favourably to those derived from other public (or public-private) sector investments. Both sets of figures are sensitive to assumptions about adoption rates among target groups. We have made more stringent assumptions about adoption in livestock because of the slow timeframe of accumulations of benefits.
Our findings are in line with the literature on returns to agricultural R&D, which demonstrates high returns to research focussed mainly on private yields. In contrast, most of the evidence here is based on the potential environmental public good returns, and greenhouse gas benefits in particular. We demonstrate how these returns can contribute to meeting at least two time bound policy targets related to the Kyoto Protocol and the Water Framework Directive. More generally the analysis suggests the extent to which plant and livestock genetics may contribute to an agriculture that makes a positive net contribution to the environment.
Our conclusions add to the suggestion in the aforementioned literature, that given high rates of return, agricultural research has generally seen systematic under-investment. This argument is all the more compelling given general scientific consensus on the rates of return that can be anticipated from further developments in life sciences, and the delivery of both environmental and health benefits from investment in animal and plant genetic research.
Our findings on the second question suggest that genetics offers a cost-effective option to address specific policy objectives. A major advantage here is that the benefits of genetic improvement are permanent and cumulative. Current investment obviates recurrent long term spending. Even though present value of these costs of alternatives can be low, they still favour a genetic approach providing adoption assumptions hold. Moreover, it is unclear how government can bring about alternative investments without potentially costly regulation.
We note that such cost-effectiveness comparisons maybe misleading because alternative interventions will typically be delivering on a basket of incommensurate benefits. Even if we restrict the analysis of effectiveness to greenhouse gas outcomes, the relative costs of implementing alternative policies (e.g. market-based instruments) are difficult to trace.
While demonstrating the potential for direct returns to prospective Defra support, our analysis abstracts from the real adoption hurdles that exist in plant and, to a lesser extent, animal science between publicly funded innovation and private sector adoption. While these hurdles exist, part of the returns presented here will remain notional. We offer some observations on how public good objectives can be advanced through research that is targeted on emerging markets for environmental goods and services (e.g. organics, biofuel and biomass).
We note that the current challenges in affecting policy through genetics are a legacy of historical reforms in plants and animal research. Specifically, both sectors are characterised by differing forms of commercialisation and cooperation, and concerns about excludability that can limit the influence of publicly funded research. Broadly speaking, ruminant livestock genetics is relatively more cooperative, with a low likelihood of small enterprises being able to capture sufficient additional market share to justify long term private research investments. Hence public funding has a more significant influence on R&D.
Private sector stakeholders are appreciative of Defra's role and the interface it offers between public research and their needs. While not always fulfilling all needs, most acknowledged the benefits of having this interface. Defra needs to focus on barriers to research adoption, and review technology transfer models that are being advocated by themselves and other partners in research funding. Researchers have a role in considering how their research delivers on public good objectives and in providing a better evidence base for their proposals.
We suggest that the evidence presented here actually understates the likely economy wide benefits of genetic research in the UK. Some of these benefits are difficult to quantify, and relate to the option value of a science base that is capable of feeding into the private sector in a variety of indirect ways.
In summary, we are confident that future investment in plant and livestock genetic research offers an economically efficient option as part of any policy portfolio to deliver public goods requirements from agriculture.
Contents
Executive summary
1.Purpose and structure of this report
1.1Introduction
1.2Background and rationale
1.3Identifying public and private roles
1.4The research process
1.5What is public value?
1.6Appraising research priorities
1.7Options appraisal
Determining priorities
Economic appraisal
1.8Measuring benefits or returns to research outlays
Returns to agricultural research and development
Impact studies related to genetic improvement research
2.Determining objectives
2.1Introduction
2.2Initial screening
2.3 Matching research with objectives
3.Web based survey
3.1Introduction
3.2Survey results
Summary of survey responses
Importance of research themes
Overall assessment of priorities
4.Economic analysis of research priorities
4.1Introduction
4.2Defining relevant policy targets
Climate change
Water Framework Directive
Other policy commitments
Organic targets
Biofuel targets
4.3Economic analysis
4.4Incremental cost-effectiveness analysis
4.5Rate of return analysis on selected research projects
4.6Livestock
Method
Survey results on livestock genetics research projects
Model of gene flow through the industry
Unit economic benefits of farm productivity changes
Unit economic benefits of greenhouse gas emissions
Unit economic benefits of reduced disease burden
Farm gate implications
Animal welfare implications
Timeframe of impact
Calculations
Results and Discussion
Conclusions
4.7Plants
Baseline
Costs
Benefits
Adoption Profile
Deriving rates of return
4.8Discussion
4.9Counterfactuals
4.10Reducing greenhouse gas emissions – comparing genetics to alternative approaches
Livestock emissions
Plant emissions
Incremental cost effectiveness analysis
5.Links between Defra funding and private sector research structures
5.1Introduction
5.2Livestock science
5.3Plant science
5.4General observations
6.Conclusions
7.References
List of annexes
Annex 1.Livestock and plant breeding industry structures
Annex 2.Main policy and appraisal statements documents
Annex 3.Project list and scoring criteria tables
Annex 4.Project Prioritisation Results
Annex 5.Example Screen Shots of the Web Based Survey
Annex 6.Summary of responses from the web based surveys
Annex 7.External costs and benefits arising from agriculture
Annex 8.Structure of livestock genetics benefits assessment model
Annex 9.Experts in discussion groups and respondent institutions to electronic survey
Annex 10.People contacted for stakeholder section
Annex 11.Original Research Specification Documents
List of tables
Table 1.1:Previous UK Studies into Returns to Agricultural R&D
Table 2.1:Outline of Defra Policy Objectives
Table 2.2:Identification of Defra Policy Goals from the Expert Groups
Table 2.3:Animal Research Areas and Projects
Table 2.4:Plant Research Areas and Projects
Table 4.1:WFD detailed timetable for implementation
Table 4.2:Summary of mean survey results for 4 different research projects with a view to influencing the rate and direction of genetic progress in UK livestock industries
Table 4.3:Discounted cumulative benefits (£000’s) from a 1% increase in performance characteristics after 10 years with 100% adoption for 5 sectors of the UK sheep and suckler beef industries using a discount rate of 3.5%.
Table 4.4:Present values of research costs and benefits, and net present values (£000’s) for 4 projects targeting improvements in the rate and direction of industry genetic progress in sheep and suckler beef breeding programs
Table 4.5:Outline of Major Parameters Adopted within the Modelling Process
Table 4.6:Median rates of yield increase and nitrogen input reduction resulting from plants genetics projects.
Table 4.7:Rates of Return and price variance (50% lower prices), for Crop Projects, percentages
Table 4.8:Non-genetics approaches to reducing agricultural GHG emissions.
Table 4.9:Comparison of genetics research with alternative approaches to methane reduction in dairy cattle.
Table 4.10:Comparison of genetics research with alternative approaches to N2O reduction in arable farming.
Table 4.11:Incremental cost effectiveness ratios for livestock genetics and counterfactual approaches
Table 4.12:Incremental cost effectiveness ratios for plant genetics and counterfactual approaches (oilseed rape).
Table 4.13:Incremental cost effectiveness ratios for plant genetics and counterfactual approaches (wheat).
List of figures
Figure 1.1:Outline of research process.
Figure 1.2:A Model of the Research Process with Agricultural Science
Figure 1.3:Defining stages of research planning
Figure 3.1: Mean importance ratings of the research themes from the livestock survey.
Figure 3.2: Mean importance ratings of the research themes from the plants survey.
Figure 3.3: Assessment of research priorities for livestock
Figure 3.4: Assessment of research priorities for plants
Figure 4.1:Framework for economic analysis.
1
1.Purpose and structure of this report
1.1Introduction
This report presents the results of Defra project ‘The rationale for Defra investment in R&D underpinning the genetic improvement of crops and animals’ (IF0101/CTX0512).
The project had two main objectives: first to help identify the elements of future research in both plant and animal genetics that most closely meet Defra’s policy objectives; second, to determine the relative efficiency of public spending on these priorities using a cost-effectiveness or a cost benefit criterion. These elements contribute part of the evidence base that could enable Defra to decide on the merits of genetics as part of its future research and policy portfolio.
More specifically the objectives were to:
- Identify where, and what, research underpinning farm species genetic improvement can contribute to Defra’s objectives
- Identify the most cost-effective alternative or currently implemented means to achieving the same outcomes, including other policy tools.
- Conduct an incremental cost-effectiveness analysis comparing additional research to the best or currently implemented intervention for achieving specific outcomes , that will take into account the timescale for delivery and Defra time-bound targets
- From the cost-benefit analysis, assess whether further investment in genetic research would be an efficient addition to existing policy instruments
- Analyse the wider deliverables of Defra-funded R&D to underpin genetic improvement, and identify the implications of a significant reduction in Defra’s investment in this area.
The project used qualitative and quantitative methods to gain an understanding of both priorities, and the economic efficiency of key genetic research for enabling Defra to meet time bound policy objectives. This analysis involved input from both UK and international stakeholders. Economic efficiency is defined in terms of whether genetic investments yield a respectable economic rate of return over the lifetime of the public investment, or are demonstrably more cost-effective relative to any feasible alternative approach.
This report is organized into six sections. Section one outlines the rationale and methodological background, and considers the public role in plant and animal genetic research as a context for conducting this study. The section then briefly reviews the evidence on the returns to agricultural R&D, and genetic research in particular. With a view to undertaking a forward looking appraisal, the section then considers the methodologies available to prioritise research, and to answer the key questions posed by this research. This requires us to consider the literature covering research priority setting and the methods for subsequently appraising the identified priorities.
Section two addresses Defra policy priorities, and how these priorities can be met by genetic research. As a basis for the economic analysis, we provide a preliminary indication as to what the SAC team has identified as headline goals for publicly funded research. With the help of input from expert group workshops, we then identify the candidate research that matches the broad policy objectives.