Introduction

The 2005 Fischler reform marked a significant milestone in the history of the common agricultural policy (CAP) of the European Union (EU), transforming the landscape by forging a direct connection between CAP payments and minimal environmental standards through the implementation of cross compliance. This reform was a proactive response to the recognition that agriculture, as a key economic sector, needs to align with sustainable practices to address environmental concerns. Cross compliance, a cornerstone of this reform, imposed a set of obligations on farmers, compelling them to adhere to statutory management requirements (SMRs) that covered a spectrum of critical aspects, including environmental preservation, food safety, public health, and animal welfare. In addition to SMRs, farmers were mandated to comply with good agricultural and environmental conditions (GAECs), outlining fundamental regulations for effective farmland management.

The overarching objective of cross compliance regulations was to instill a sense of responsibility among farmers, who, as beneficiaries of subsidy payments sourced from the wider European Union taxpayer base, were expected to deliver tangible environmental benefits in return for the financial support they received. This innovative approach aimed to reconcile agricultural practices with broader environmental imperatives, recognizing the pivotal role of farmers in fostering sustainability within the sector (Bartolini et al. 2012; Bennett et al. 2006; Hart et al. 2012).

Since cross compliance was first introduced, the CAP has undergone a series of reforms reflecting the evolving priorities and challenges faced by the agricultural sector. The most recent reform, as negotiated by representatives from the European Parliament, the Council of the European Union, and the European Commission, underscores a concerted effort to propel the CAP into a new era—one characterized by sustainability, competitiveness, and resilience. The new CAP promises to be “A greener and fairer CAP” and has a budget of €387 billion (2021–2027), with an earmarked €8 billion dedicated to supporting rural areas in making structural changes aligned with the goals of the European Green Deal and digital transition (EU Commission 2021).

In line with the objectives of fostering a greener CAP, European farmers are now faced with heightened environmental conditionality, where the receipt of payments is intricately linked to a more robust set of mandatory requirements. The success of the newly implemented CAP measures, designed to marry agricultural productivity with environmental stewardship, hinges on the acceptance and adherence of farmers to these new regulations. This paper embarks on a comprehensive exploration of farmers’ perspectives on the linkage between environmental conditions and subsidy payments, by looking at the attitudes and perceptions of farmers to linking direct payments to compliance with environmental regulations. Understanding farmer attitudes toward cross compliance is important due to its far-reaching implications for policy effectiveness, voluntary adherence, and the enduring sustainability of agricultural practices.

The research questions guiding this exploration are threefold. Firstly, we will investigate the degree of farmer endorsement for the connection of CAP direct payments with cross compliance and good agricultural practices. This entails understanding the extent to which farmers perceive the link between their financial incentives and the adoption of environmentally sustainable practices as beneficial or burdensome. Secondly, the paper aims to shed light on the evolution of farmer perspectives on environmental conditionality over time. By tracing the trajectory of attitudes, we seek to discern patterns, shifts, and factors influencing the dynamic relationship between farmers and environmental regulations. Lastly, we will scrutinize the determinants that influence the acceptance or rejection of the principle of linking payments to environmental conditions. This involves unraveling the intricate interplay of factors such as economic considerations, social dynamics, and individual beliefs that shape farmers’ responses to the changing policy landscape.

Previous literature

The introduction of environmental conditionality, as exemplified by cross compliance, signaled a shift in the requisites for goods and services supplied by land managers in exchange for financial assistance from European taxpayers (Latacz-Lohmann and Hodge 2003). Existing literature concerning cross compliance has generally concentrated on two main areas: the potential repercussions of heightened regulations on farm business and the capacity of cross compliance to fulfill the broader societal demand for environmental public goods (Davies and Hodge 2006). In this paper, we focus on the former and investigate the influence of socio-economic factors on farmer attitudes toward linking direct payments with enhanced environmental conditions in the form of cross compliance.

Traditional command and control policies, such as cross compliance, establish standardized practices and regulations across the agricultural sector ensure consistency and quality in production, contribute to food safety and consumer confidence, and often include measures to protect the environment (Holling and Meffe 1996). However, there are a number of difficulties associated with this kind of regulation. Literature highlights that some of the difficulties of command and control policies include the imposition of significant costs on farmers and a high level of regulatory burden (Lambin et al., 2014; Sinclair 1997). There may also be unintended consequences associated with this type of regulation, such as the displacement of environmental impacts to other regions or sectors or a lack of recognition of the complexity and the interconnectedness of the natural world (Suckling et al., 2021). From an efficiency point of view, command and control may hinder innovation, or prevent farmers from adapting to local circumstances or natural constraints (Bareli et al., 2020). The lack of flexibility in command and control policies can stifle agricultural development and productivity improvements and fail to address difficult environmental problems such as water pollution (Sharma 2020).

Past studies have also indicated that the effectiveness of command and control policies, such as cross compliance, hinges on a combination of effective and cost-efficient monitoring by the governing body and widespread compliance within the farming community (Nitsch and Osterburg 2008). Preconditions for high levels of farmer compliance include knowledge and awareness of the rules and acceptance that the rules are fair and reasonable (Winter and May 2001). According to Nitche and Osterburg, the main objective of enforcement mechanisms should be to strengthen the trust of citizens in “fair arrangements.” Therefore, we would expect farmer attitudes towards cross compliance to adapt over time, and where rules are deemed fair and reasonable, higher levels of compliance would reflect this opinion. However, despite ongoing efforts by the regulating authorities, there continues to be a significant level of non-compliance among some Irish farmers (Lunn et al., 2020).

Within the Irish context, Lunn et al. (2020) employ administrative data from the Department of Agriculture to undertake a comprehensive statistical analysis to investigate the drivers of non-compliance with the Nitrates Directive. Their analysis focused on the attributes of farms and farmers exceeding the regulatory threshold of 170 kg of nitrogen per hectare (N Ha−1), excluding farms that had obtained approval for an exemption to operate beyond the 170 kg limit and farm up to 250 kg N Ha−1, known as a derogation. Their findings suggest that previous farming decisions were important and that farms operating at higher levels of N Ha−1 in previous years were at an increased risk of non-compliance. Farms that had previously exceeded the limits were more likely to do so again, despite the risk of penalties. The study revealed that farm size emerged as a notable predictor of non-compliance, with smaller farms demonstrating a higher likelihood of violating the regulations. The age of the farmer also proved to be a significant factor, as older farmers were found to have a lower likelihood of non-compliance. Combining these findings, the authors assert that regulatory breaches are most likely to happen during periods of change within a farm business, potentially attributable to factors such as adopting a new business model, scaling up production, or the transfer of land ownership (Lunn et al., 2020).

Davis and Hodge (2006) examined the perceived legitimacy of cross compliance as a governance mechanism among farmers. They discovered that two key attitudinal factors, labeled as "stewardship orientation" and "technological beliefs," were overwhelmingly the most significant factors in determining the acceptability of cross compliance (Davies and Hodge 2006). The findings from a Q methodology conducted with Irish farmers in 2012 indicated an increasing acknowledgment and acceptance among farmers regarding the environmental advantages of the Nitrates Directive. However, skepticism persisted concerning the validity of certain measures (Buckley 2012). While individual motivations and attitudes are important factors in farmer acceptance of environmental regulations, social norms were found to be as important as individual motivations (Prager and Posthumus 2010).

From an economic perspective, financial rewards were found to be a positive influence on farmer acceptance of cross compliance (Posthumus et al., 2011; Merckx et al., 2009; Mills et al., 2017a). However, financial benefits alone are seen as transient drivers without long-term stability (Mills et al., 2017a). Increased transaction costs, on the other hand, whether actual or perceived were found to have a negative effect on farmer acceptance of environmental conditionality. Transaction costs may take the form of administrative, information, and organizational costs, and all add to the workload of the farmer and may involve additional skills which the farmer may not be familiar with (Zinngrebe et al., 2017). This increase in administrative burden may be viewed differently by farmers, and well-educated and well-informed farmers have been shown to have a more positive attitude toward agricultural policy and perceive administrative tasks as less onerous (Ritzel et al., 2020).

Farmer attitudes and beliefs were also found to be significant determinants of farmer acceptance of cross compliance (Morris and Potter 1995; Defrancesco et al., 2008; Wauters et al., 2010). Attitudes can be defined as a mindset or a tendency to act in a particular way due to both an individual’s experience and temperament. Attitudes are a complex combination of personality, beliefs, values, behaviors, and motivations (Pickens 2005). Attitudes can change but transformation takes time, effort, and determination.

Research has demonstrated that positive farmer attitudes correlate with higher levels of compliance, thereby enhancing the overall effectiveness of cross compliance policies (Beedell and Rehman 1999; van Dijk et al., 2016; Bartolini et al., 2021). These attitudes also influence the willingness of farmers to adopt environmentally sustainable practices, playing a crucial role in achieving the environmental goals outlined in cross compliance (Menozzi et al., 2015; Elahi et al., 2022). Farmer endorsement of cross compliance not only leads to active support and engagement but also facilitates effective communication between policymakers and the farming community (Mills et al., 2017b).

This study enhances the current literature by directly seeking farmers’ opinions on the incorporation of environmental regulations into CAP pillar one direct payment. We examine farmer attitudes toward cross compliance across two distinct periods, specifically in 2010 and 2018. This allows us to examine potential shifts in farmer perspectives over time and assess what factors influence changes in these attitudes.

Conceptual framework

The theoretical underpinning of this paper is based on a multidimensional intertemporal utility framework. Multidimensional intertemporal utility integrates two critical dimensions—multidimensionality, recognizing that farmers derive utility from various aspects of their lives, while also acknowledging the evolving nature of preferences over time. In this framework, farmers navigate a complex web of considerations, balancing short-term economic gains with long-term environmental stewardship. In its simplest form, a multidimensional intertemporal utility function can be expressed as

$$U=U(t,X_1,X_2,\dots,X_n)$$

where \(U\) represents multidimensional intertemporal utility, \(t\) denotes time, and \(X_1,X_2,\dots,X_n\) represent the different dimensions or attributes contributing to a farmer’s well-being. This mathematical representation captures the dynamic and diverse nature of farmers’ utility considerations.

In this paper, the dimensions of farmer utility include economic viability, environmental stewardship, and community dynamics. Command and control policies such as cross compliance, directly affect the economic viability of the farm and therefore have an influence on farmer decision making. From an economic viability perspective, farmers weigh short-term economic gains from subsidy payments against the potential long-term costs associated with cross compliance regulations and penalties for non-compliance. The environmental stewardship dimension recognizes that farmers also derive satisfaction from sustainable farming practices that contribute to environmental preservation. Within the community dynamics dimension, social relationships within farming communities help shape farmer perceptions and attitudes; hence, the influence of peers can have a positive or a negative effect on farmer’s acceptance of cross compliance measures. The intertemporal nature of the decision acknowledges that farmers must balance the immediate economic benefits of cross compliance with potential long-term gains in environmental sustainability. This involves optimizing resources over time, considering the multidimensional aspects of farming practices and compliance efforts.

This study contributes to the existing body of literature by broadening the discourse surrounding command and control policies. These policies not only overlook the intricate interconnections within the natural environment but also fail to account for the diverse human factors that significantly influence decision-making. Ultimately, these human dimensions play a pivotal role in the effectiveness or ineffectiveness of regulatory measures. The framework presented here offers a perspective for dissecting the complexities of farmer decision-making and evaluating the impacts of command and control policies on agricultural sustainability and community welfare.

Data and methodology

Data

Data for this study was collected through the TeagascFootnote 1 National Farm Survey (NFS) in Ireland, which is part of the EU Farm Accountancy Data Network (FADN). The EU FADN gathers economic data on farms across the EU for the derivation of incomes and business analysis of farm holdings (Council Regulation (EC) No 1217/2009). The Teagasc NFS is collected annually, and farms are selected on a stratified random basis in conjunction with the Central Statistics Office (CSO) of Ireland. Each farm is attributed a weighting factor so that the results of the survey are representative of the proportion of farms nationally (Dillon et al., 2017). The data is collected by a team of professional recorders and includes variables such as livestock numbers, cropping area, inputs and outputs, assets and liabilities, direct payment from CAP under pillars 1 and 2, and family farm income.

A supplementary survey was designed and administered in conjunction with the core NFS data in 2010 to assess farmers’ attitudes to linking CAP payments to environmental conditionality. Farmers were asked to express their agreement or disagreement with the statement “Farmers should only be eligible to receive CAP basic payments scheme monies if they meet good agricultural practice and cross compliance standards.” The data is presented on a Likert scale index, ranging from 1–5, where 1 corresponds to “strongly disagree,” 2 to “disagree,” 3 to “neutral,” 4 to “agree,” and 5 to “strongly agree." The same set of questions was asked again in 2018. We analyze the outcomes from these two periods to evaluate potential shifts in attitudes over time. A balanced panel of farms, combining results from 2010 and 2018, results in a final dataset for this analysis comprising 916 farms, weighted to ensure representativeness among 74,507 farms nationally.

Methodology

The dependent variable for this study is categorical, in which the order of the response is important therefore an ordered logit model would seem appropriate. In ordered logit models, or proportional odds models, the slope coefficients are assumed the same for each category; it is only the intercepts or the cut-off points that differ. Therefore, one of the assumptions inherent in ordered logistic regression is that the relationship between each pair of outcome groups remains consistent. In other words, ordered logistic regression assumes that the coefficients describing the relationship, for instance, between the lowest category versus all higher categories of the response variable, are identical to those describing the relationship between the next lowest category and all higher categories. This assumption, known as the proportional odds assumption or the parallel regression assumption, is frequently breached in practice.

To investigate the factors influencing farmer attitudes toward cross compliance, we employ a generalized ordered logit model (Gologit). This model determines the odds of being in a particular category relative to the preceding one, providing advantages over the normal ordered logit model by alleviating the necessity for the parallel lines assumption (Williams 2016).

The original ordered logit model is defined as

$${Y}^{*}= {B}_{1}{X}_{i1}+ {B}_{2}{X}_{i2}+\dots {B}_{k}{X}_{ik}+ {u}_{i}$$

where \({Y}^{*}\) is the outcome (1 to 5) to the question “Farmers should only be eligible to receive CAP basic payments scheme monies if they meet Good Agricultural Practice, Cross compliance standards”, X are the independent variables, and \({u}_{i}\) is the error term.

Farmers face \(j\) ordered alternatives with cut-off or threshold points \(a\) such that

$$Y_i=1\;if\;Y_i^\ast\leq a_1$$
$$Y_i=2\;if\;a_1\;\leq\;Y_i^\ast\leq a_2$$
$$Y_i=3\;if\;a_2\;Y_i^\ast\leq a_3$$
$$Y_i=j\;if\;a_j-1\leq Y_i^\ast$$

where \({a}_{1} < {a}_{2} < {a}_{3}\dots . <{a}_{j-1 }\)

There are five \(j\) alternatives: strongly agree, agree, neither agree nor disagree, disagree, and strongly disagree ordered 1–5; subsequently, there are four cut-off or threshold points \(a\).

The probability that a farmer chooses j is defined by the odds ratio:

$$\frac{\text{Pr}\left[{Y}_{i} \le j \right|X]}{\text{Pr}\left[ {Y}_{i}>j |X\right]} = \frac{\text{Pr}\left[{Y}_{i} \le j \right|X]}{\text{Pr}\left[ 1-\text{P}\text{r}({Y}_{i}>j |X)\right]}$$

where \(\text{Pr}[ {Y}_{i} \le j \left|X\right]= \sum _{m=1}^{j}\text{Pr}[ {Y}_{i}=m \left|X\right]\) is the cumulative probability that the outcome is less than or equal to j. Taking the log of this odds ratio gives

$${\mathrm{Logit}\left[\text{Pr}\left(Y_i\leq j\right)\right]}=\text{ln}\frac{\text{Pr}\left(Y_i\leq j\right)}{\left[1-\text{Pr}\left(Y_i\leq j\right)\right]}$$

Should the parallel line assumption prove accurate, the coefficients of the logistic regression are expected to exhibit consistency across every category. Similarly, under the assumption of proportional odds, the odds ratios should remain constant for each ordered category of the outcome variable.

The generalized ordered logit model is more restrictive than the ordinal logit model, choosing to relax the assumptions of the ordered logit model only when deemed necessary. The Gologit model can be written as

$$P\left(Yi > j\right)= g\left(X\beta \right)= \frac{\text{exp}\left( {\alpha }_{j+ }{X}_{i }{\beta }_{j }\right)}{1+\left\{\text{exp}\left( {\alpha }_{j+ }{X}_{i }{\beta }_{j }\right)\right\}} , j=\text{1,2}, \dots , M-1$$

where αj are the intercept, βj are the logit coefficients, and M is the number of categories of the ordinal dependent variable, in this paper M = 3. From the above, it can be determined that the probabilities that farm \(i\) will fall into any category \(j\) from 1, …, M are defined as follows

$$P\left(Yi= 1\right)=1- g\left({X}_{i }{\beta }_{j }\right)$$
$$P\left(Yi= j\right)= g\left({X}_{i }{\beta }_{j-1 }\right)- g\left({X}_{i }{\beta }_{j }\right) j=2 , \dots , M-1$$
$$P\left(Yi= M\right)= g\left({X}_{i }{\beta }_{M-1 }\right)$$

This model estimates the odds of being in a certain category relative to being in the previous category and provides empirical means to identify which variable violates the parallel lines assumption of the Gologit model.

Results

Over the specified period, there has been a strong level of support among this cohort of farmers for linking direct payments to good agricultural practice and cross compliance standards (Fig. 1). This implies that from the beginning, there was a favorable attitude toward recognizing and supporting the link between adopting sound agricultural practices and fulfilling cross compliance criteria within the farming community. The average agreement score on a scale of 1 to 5 increased from 3.9 in 2010 to 4.1 in 2018. In 2010, 71% of farmers expressed either agreement or strong agreement with the statement, and by 2018, this percentage increased to 82%. Conversely, the proportion of farmers who disagreed or strongly disagreed with the statement decreased from 11% in 2010 to 3.9% in 2018. The change in farmers’ perspectives on the notion of environmental conditionality indicates a positive trend, highlighting their acknowledgment of the role they play in offering public goods through agri-environmental practices.

Fig. 1
figure 1

Farmer support for linking direct payments to cross compliance conditions 2010 and 2018

Full size image

A two-sample t-test was performed to compare the 2010 findings with those of 2018, investigating the hypothesis that there is no significant difference in farmer responses between these two distinct years (Table 1). Monitoring changes over time becomes more straightforward when every unit is consistently observed in all periods; therefore, the panel is strongly balanced.

Table 1 Two-sample t-test with equal variances
Full size table

The findings presented in Table 1 highlight a substantial difference in the responses provided by farmers in the years 2010 and 2018; therefore, we cannot accept the hypothesis that there is no difference between these two years. This positive outcome points toward an observable trend over this period, indicating that farmers are displaying a diminishing reluctance toward environmental conditionality. Specifically, they are becoming less inclined to express disagreement with the idea of associating their payments with cross compliance and adhering to good agricultural practices, and the number of farmers who strongly disagreed or disagreed neutrally declined between the two periods. On the other hand, the number of farmers who strongly agreed also fell during this period, but overall, the number of farmers who agreed with linking payments with environmental outcomes increased. This shift in attitudes suggests a growing recognition or acceptance within the farming community regarding the importance of aligning agricultural activities with environmental standards and regulations. It reflects a positive evolution in farmers’ perspectives, potentially indicating an increased understanding of the broader environmental considerations associated with their practices and the link between financial support and sustainable farming practices.

The subsequent phase of this analysis involves investigating the factors contributing to the observed shift in attitude among Irish farmers. To explore these influences, we employ a generalized logit model, as detailed in the "Previous literature" section. The outcomes of the generalized ordered logit model are outlined in Table 2, wherein each category is individually assessed for differences relative to the preceding category. This comprehensive examination allows us to discern the nuanced impact of various factors on farmers’ attitudes, providing insights into the drivers of the evolving perspectives within the agricultural community.

Table 2 Results of generalized ordered logit model agreement with GAP and cross compliance link to CAP payments
Full size table

The results indicate that as we move across the spectrum of farmers who strongly disagree to farmers who strongly agree with the policy, farm size, farmer age, the level of payments received, and how dependent the farm is on those payments are all important predictors of farmer responses. Since decoupling, farm direct payments are not linked to production, but rather to past production decisions in which larger farms received larger payments based on the level of production (livestock numbers or area under crops); therefore, larger farms were likely to have larger payments. In this analysis, farmers with larger farms and larger direct payments Ha−1 are more likely to agree with the compliance regulations. However, these farms are also less dependent on direct payments, which indicates that these farms are making a profit from the market, since the direct payments make up a significantly less portion of their overall farm income. While larger farms may be more amenable to compliance regulations, their reduced reliance on direct payments highlights their capacity to thrive within market dynamics, thereby influencing their perspectives on regulatory measures.

To put this in reverse order, farmers who disagree with the policy are more dependent on direct farm payments, and their overall farm income is more at risk if their farm is found to be non-compliant. The farmers who disagree are also more likely to have dry stock farms, which include cattle and sheep farms. Historically, these farm systems have also been highly dependent on subsidy payments. European FADN data shows these farms to be low-income enterprises, and in some cases, they are using direct payments to subsidize farm production, i.e., where the direct payments are over 100% of farm income (Martinez Cillero et al., 2019; Taylor et al., 2018; Regan et al., 2018). This result follows the previous result where farmers who were more dependent on direct payment disagree with the policy, and the farm systems that are more dependent are cattle and sheep farms. The identification of dry stock farms as particularly vulnerable highlights the sectoral disparities within agriculture and the uneven impacts of command and control policies on different farming systems. This underscores the importance of considering sector-specific challenges and vulnerabilities when designing and implementing regulatory frameworks.

Factors identified in the analysis that influence farmer attitudes toward cross compliance can be categorized under a number of different headings, which include both internal and external factors. These categories are listed in Table 3.

Table 3 Factors identified in the analysis and categories they fall into
Full size table

Discussion and conclusion

As farmers continue to adapt to policy changes in the CAP and acknowledge their role in environmental protection, policy in relation to the environment continues to evolve. The level of farmer compliance with new environmental conditions will have implications for farmers and the wider society. High levels of compliance also reduce the cost of enforcement; therefore, investigating the factors that influence compliance with existing regulations is important. If individuals feel that the rules and the requirements under new policy regulation are fair, then we would expect a high level of compliance; however, if individuals feel that the policy rules are impractical and unfair, then we would expect a higher instance of non-compliance.

In this paper, we investigated Irish farmers' attitudes toward environmental conditionality in the form of cross compliance, to identify the socio-economic factors that influence these attitudes. Results indicate that the higher proportion of subsidy payments in the overall farm income is a significant indicator of farmer disagreement with the policy. The risk associated with non-compliance and the possibility of fine and loss of income is not the same for all farms, farmers who disagree face a higher risk associated with non-compliance, since subsidy payments constitute a larger share of their farm income. This is a problem often associated with the one-size-fits-all approach of command and control policies, where all farmers must comply with the same rules but the loss of income associated with non-compliance has a different effect on the overall profitability of farms more reliant on the payments.

The high level of support for linking CAP payments to good agricultural practice and cross compliance standards is also an indication that farmers realize they have a role to play in protecting the environment and in protecting the future sustainability of the sector. However, as a caveat, agreement with the policy does not necessarily indicate future compliance. Factors that influence farmer acceptance of cross compliance can be categorized under a number of different categories, farm and farmer characteristics, social/institutional characteristics, economic characteristics, and decision or choice characteristics. All of these different influences on farmer attitudes toward the principle of cross compliance are an indicator of both the complexity and the level of heterogeneity at the farm level and the number of internal and external influences that are at play in forming attitudes toward the policy.

To investigate if attitudes toward cross compliance have changed over time, we compared responses from 2010 to 2018. Our results show that there was a high level of agreement with linking direct payments to cross compliance regulations in both periods. There was also a significant difference in the overall disagreement with the policy measures, and fewer farmers disagreed with the policy in 2018 than in 2010. This is a positive result and indicates that there is less resistance to the policy over time and farmers accept the important role they play in protecting natural resources and reaching national environmental targets.

A positive shift in farmer attitudes not only reflects an increased willingness to embrace environmental measures but also underscores a fundamental adaptability within the agricultural sector to confront and navigate evolving environmental conditions. This adaptability holds critical implications for addressing emerging environmental challenges and adapting to the dynamic nature of policy frameworks. From a wider policy perspective, it may encourage policymakers to innovate and evolve environmental policies based on the feedback and cooperation of the communities involved. This iterative process can lead to more effective and targeted policy measures. The feedback loop allows policymakers to respond to real-world challenges, ensuring that policies are aligned with the needs and perspectives of those implementing them. Policymakers can leverage this information to identify successful strategies and incorporate them into policy frameworks, fostering the dissemination of effective and sustainable practices.