The team from ESSRG, overseeing the policy work package in Contracts2.0, conducted a Delphi-survey to highlight limitations of the current agri-environmental schemes as well as to gauge the perception of innovative approaches for the delivery of public goods amongst policy makers, farmers/advisors, researchers and NGO’s across Europe.

The Delphi method is based on structured expert surveys and draws on experiences and knowledge of the participants in form of empirical, predictive and normative aspects. Its core concept is to facilitate discussions and develop consensual ideas among participants via several correlated rounds allowing experts to reconsider their opinion in each round.

It is planned to run altogether three rounds of the Delphi survey, to assess the knowledge base and to stimulate a discussion regarding the adoption of innovative contractual solutions for a farmer- and eco -friendly agriculture. The first round of the survey, conducted in March/April 2021, was completed by 41 stakeholders form 17 European countries. The aim was to first depict the limitations of current agri-environmental schemes, followed by an assessment of the knowledge base and experiences regarding the four novel approaches being at the heart of contracts2.0: results-based payments, collective action as well as value chain and land tenure approaches. In the following some of the results are summarized:

Limitations of the current schemes are evident

Regarding the limitations the results suggest, that financial (e.g. transaction cost, compensation vs. reward) and institutional (e.g. bureaucracy, lack of flexibility) factors are the most relevant. But knowledge-related aspects (e.g. lack of robust scientific evidence, limited access to advisory services) as well as the need for improved monitoring and regional differentiation were also mentioned as hindering an effective implementation of agri-environmental measures.

Results- vs. action-based payments

Concerning the innovative approaches, results-based payments were perceived as potentially (highly) effective in achieving the ecological objectives (especially biodiversity related) due to the agreement of clear and measurable outcomes. Respondents highlighted that rewarding farmers for their environmental performance (instead of compensating for their lost income or increased costs) contributes to the attractiveness of this approach. While the increased flexibility and autonomy for farmers are additional advantages, issues like monitoring (e.g. definition of robust indicators, use of IT or farmers expertise to bring down cost) and risk mitigation (uncertainty due to external factors) still pose a challenge and need some further attention when refining/adapting this approach. Many respondents state that setting up such schemes would require initially large investments (management, monitoring, trainings), which might serve as a barrier for the adoption. As a potential middle ground some experts suggest a combination of results-based payment as a bonus on top of an action-based implementation to reward more ambitious efforts.

Collective action

Many respondents agreed that the collective approach (or group contracts) can effectively deliver on the (mostly biodiversity related) objectives when adequate ecological expertise is involved. The main argument in favor of cooperation is the positive effect on the connectivity of habitats through a coordinated management of suitable measures on landscape scale. The decreased (real or perceived) transaction cost on the farmers’ side is another advantage. While some respondents argue a long-lived tradition with cooperation amongst farmers helps to succeed when collectively implementing measures, others highlight the importance of a feeling of trust within a collective and/or contractual elements for regulating individual behavior being the key factors for success. Regarding the transaction cost of the collective approach respondents acknowledge a shift from public transaction cost (administrations) to the private sector (mainly the collective itself or their management respectively).

Value chain approach

The value chain approach is well received by most respondents, since it seems well suited to reward farmers for their environmental performance, independently from public funding. The generation of income through an adequate market price is also more in line with farmers’ business logic (instead of relying on public funds). This approach seems to work well with shorter value chains and a more regionalized marketing, while large retailers do often not engage so easily in such programs. A key factor of this approach is the labeling of the extra effort to simplify the decision to purchase for the environmentally conscious customers. Given the already large variety of labels, special attention needs to be paid to a transparent and clear communication of the environmental performance throughout the value chain. Respondents recognize no significant (institutional, cultural or social) barriers to implement this approach, however it is perceived to require an extensive infrastructure and a broad knowledge base.

Land tenure contracts

The Land tenure approach is the one with the most uncertainty among the respondents, possibly due to a potential lack of experience with this type of contract. It could be a well-suited approach for large landowners, especially when intermediaries (National Parks, Church, NGO’s) are involved. The comparatively longer contract periods of land-tenure approaches not only contributes to the environmental effectiveness but also benefits the farmer by provided a sound and predictable financial base, where uncertainty and risks are comparatively low.

How does the ideal contract look like?

The last block of the survey addressed the question of the ideal contract. Interestingly the (assumed) most effective type of contracts across all respondents were by far the publicly funded agri-environmental scheme (AES). Regarding the characteristics of a contract the opinions differ a bit more. Bi-laterally agreed and publicly funded contracts with a flexible length and a mix of results- and action-based payments earned the highest approval rate. Although the results of this survey are far from being representative, the glimpse into the minds of different stakeholder groups involved with the discussion of agricultural policy issues, has provided us with some valuable insights. Especially the open-ended questions have generated a rich pool of ideas, to guide us on our path towards the development of contractual solutions that benefit farmers and nature. The third and last round of the Delphi will be completed in September 2021.

For a more detailed and comprehensive interpretation of the results check out this report.

Figure 2. Differences of the characteristics of the ideal contract according to the background of the respondents. Source: own compilation based on the 1st round of the Delphi survey.