Final Report for GS06-055
Project Information
Agricultural policy is becoming increasingly influenced by environmental policy, but achieving environmental objectives through government conservation programs remains an elusive goal. Numerous studies have shown that farmers are motivated to adopt conservation practices not only by profit incentives, but also by factors such as perceptions of good stewardship and attachment to the land. In this analysis, I use innovation-adoption theory to analyze case studies of crop farmers in the Lower Roanoke watershed in North Carolina to show how they are influenced by micro-level dynamics at the farm scale. Using this theory, I explain how farmer perceptions of conservation practices can both encourage and discourage them to adopt these practices. This analysis shows that practices that provide a perceived relative advantage, through things like labor and time savings, are more readily adopted, but practices that are perceived as incompatible with farmers’ values, needs, or relationships with landowners are frequently rejected, even when economic profits might be derived from them. I also found that farmers are strongly influenced by biophysical aspects of their farm operation and have varying, but influential, interpretations of what good stewardship looks like. Based on these findings, I suggest that conservation programs could be marketed in a more effective way by appealing to farmers’ motivations and by communicating conservation benefits in ways that take farmer perceptions into account. In particular, conservation programs should be framed in a way that educates farmers about how conservation can benefit the quality of their land, crop yields, and agroecosystem.
Introduction
Increasingly, federal support for agriculture is being directed toward conservation programs, and many policy analysts are predicting that environmental objectives will guide the future of agricultural policy in the United States (Napier et al. 2000; Batie and Horan 2002; Heimlich 2004; Claasen and Morehart 2006). The United States Department of Agriculture (USDA) has promoted conservation practices, largely through incentive payments, since the Dust Bowl era. However, widespread concern about agricultural impacts on water, wildlife, and soil quality persist. A number of researchers have argued that government programs targeting conservation on farmland have been inefficient and have failed to meet ecological objectives (Feather and Amacher 1994; Forster 2000; Hellin 2006). In the proceedings from a 1996 international conference on soil and water conservation (Napier et al. 2000), authors call for new policies that address the increasing social, political, and economic complexities of agriculture in the new millennium. A central component of this discussion is how to make conservation programs more effective, despite the widespread heterogeneity of farmers’ needs, attitudes, and capacities.
Understanding what motivates farmers to adopt conservation has been the subject of several decades of research. A literature review identified a wide array of findings about conservation adoption behavior from public policy, economics and sociology research. Many of these studies focus on understanding and projecting conservation adoption using quantitative analysis of sample survey data, correlating farm structural variables (such as farm size, age of farmers, etc.) with information about farmer behavior (Blase 1960; Ervin and Ervin 1982; Fuglie and Kascak 2001).
A 2006 study by the USDA’s Economic Research Service that examined trends among a national sample of farmers, showed that practices that are compatible with economic profits are adopted by all types of farmers and not limited to a certain group (Lambert et al. 2006). Feather and Cooper (1995) found that economic profits might be the most important motivator, but showed that awareness of conservation benefits and knowledge about how to use an innovation were equally important. Chouinard (2006) used a combined stewardship-profit motive benefits framework to examine farmer willingness to trade profits for stewardship and found that farmers were willing to trade. A recent study of Midwestern farmers (Ryan et al. 2003), found that variables such as farm income, farm size, age of farmer, and other common estimators were not significant explanatory variables of farmer motivations to adopt conservation and that attachment to the land was the most significant variable.
It is clear from these contradictory findings that adoption motivation is complex and not fully understood. Napier et al. (2000) suggests that using socioeconomic, farm structure, and public policy variables to predict adoption is problematic, pointing out the contradictions in the literature and presenting new findings from a large study of Midwestern farmers where many of the standard variables, including the receipt of government incentive payments, did not correlate with adoption behavior. Napier recommends that social scientists focus on composite conservation adoption behavior (or a comprehensive analysis incorporating all aspects of farmer behavior relating to conservation on the farm), as opposed to focusing on the adoption of specific practices, in order to understand the process of adoption. He recommends more research at the farm scale to flesh out what influences farmer behavior, suggesting that the existing conservation programs may be ineffective because they are based on incorrect assumptions about why farmers decide to adopt conservation practices.
Amidst this growing body of literature there is a paucity of case studies illustrating composite conservation attitudes and behavior at the farm level. Combining qualitative insights with the depth of quantitative studies that have previously been conducted can shed light on the complexity of real-world situations on farms. My research hypothesizes that farmer attitudes and behavior interact in ways that cannot be observed in large-scale surveys and that these micro-level behaviors may be even more important than the standard variables generally used in sample surveys to determine the adoption of conservation practices.
Understanding micro-level farmer behavior is particularly important for the design of new federal “green payment” programs that are intended to provide income support, along with an incentive for conservation. The Conservation Security Program, a new provision in the Farm Bill (2002) and the country’s first green payment program, was developed to “reward the best and motivate the rest” by paying farmers for conservation related to production (such as the reduction of soil disturbance during production) and by providing payments for enhanced conservation over the years of enrollment. Farmers agree to maintain their qualifying practices for 5 to 10 years as part of the contract and can elect to undertake any of a number of enhancements voluntarily. The CSP is a more flexible conservation program than previous Farm Bill programs because it provides incentive payments for environmental benefits, as opposed to trying to instigate specific practices. As a result, farm operators have the liberty to decide what innovations they want to adopt in order to provide water, air, soil, and wildlife benefits. Farmers’ choices about which innovations to undertake, if any, will determine how effective the program is at achieving environmental improvements. Furthermore, since the program was designed to reward those farmers who have already undertaken voluntary conservation practices, knowing what has motivated these “good actors” to undertake conservation prior to receiving incentives is important for policymakers to determine what will motivate others to adopt and to enhance the practices of already conservation-oriented farmers.
This research presents case studies of twelve farmers in one of the first watersheds selected for the Conservation Security Program, the Lower Roanoke in eastern North Carolina. These case studies describe what has influenced these farmers’ decisions to adopt conservation practices, how they perceive their impacts from conservation today, and what limits them from adopting additional conservation practices. This research provides policymakers with an example of why and how farmers undertake conservation, augmenting the existing literature showing who participates and what practices they implement.
1. To qualitatively assess why selected farmers have adopted or not adopted conservation practices in the Lower Roanoke Watershed.
2. To understand how farmers perceive the impacts of their conservation practices.
3. To relate findings from these 12 case studies to other research.
4. To determine whether the Conservation Security Program marked an increase in farmer innovation and/or an increase in conservation awareness in the watershed.
5. To make policy recommendations about how to increase the effectiveness of conservation programs based on the findings.
Cooperators
Research
Methodology
I used case studies to document and analyze the innovation-adoption process. The purpose of a case study is different from a sample survey design in that it does not attempt to determine the prevalence of phenomenon; rather it attempts to investigate a phenomenon in context (Yin 2003). I used a multiple case design, the purpose of which is to replicate a phenomenon in multiple scenarios (Yin 2003). For my research, I considered each individual farmer to be a unique study to which I applied the same theoretical framework of adoption-diffusion theory. I was then able to draw cross-case conclusions.
I developed a survey instrument based on background research conducted over the summer of 2006 for Environmental Defense (Henry and Lurie 2006) and through a review of the literature. Ialso consulted with experts from the Natural Resources Conservation Service in North Carolina (NRCS), non-profit organizations such as Environmental Defense and the American Farmland Trust, Duke University’s Nicholas School of the Environment and Earth Sciences, and the USDA social science research team based in Greensboro, North Carolina. These expert consultations were designed to narrow my research question to what was relevant to conservation program policies. Through these expert consultations, I learned of a deficit of information pertaining to what limits farmers from undertaking conservation practices, compared to the amount of information exploring motivations to adopt conservation practices. I also learned that there is dissatisfaction about the amount of conservation that has been adopted on agricultural lands in the Lower Roanoke watershed and frustration about the ongoing impacts from agricultural on the region’s water quality.
I held a focus group with farmers from the region prior to the implementation of the survey and adapted the questions based on their feedback. The focus group session was unstructured and farmers mainly talked about the process of considering whether or not to adopt an innovation. Participants commented that farmers in the region were losing money by not using certain conservation practices and suggested that the reason for this might be a resistance to change.
Subject Selection
I obtained the names of interview subjects from a variety of sources, including the NRCS district conservationists, USDA extension service, farmers, and farming advocacy groups. All farmers agreed to be interviewed in accordance with a Duke University-approved verbal consent agreement.
I grouped respondents into two categories: recipients and non-recipients of Conservation Security Program funding. I interviewed six individuals from each category. In each category, for comparative purposes, I categorized farmers based on USDA farm characteristics (table not included).
I conducted semi-structured interviews with each farmer, using a set of research questions based in adoption-diffusion theory. Interviews lasted anywhere from forty-five minutes to four hours, depending on the farmer’s interest in speaking more about the research questions. However, the same protocol was followed regardless, with extra time used for probing further into some of the farmer’s answers to the general set of research questions and to touring the farm. Questions were open-ended, allowing farmers to come up with responses unprompted (as opposed to pre-defining terms such as “conservation”). Probing was used to get farmers to explain their answers and to fully explore their opinions.
I recorded interviews (with each farmer’s permission) on an iPod for later transcription. I was the only one to conduct interviews, allowing me to minimize response bias by using the same approach to questioning with each interview subject. Response bias may have resulted from my association with an environmental management program, however I made clear in my introduction that responses would be kept strictly confidential, that there were no right answers to any of my questions, and that I wanted their honest opinions about these topics, since their input would provide valuable insight about what farmers really care about when it comes to conservation. Considering that all contacted farmers who were available during the two weeks during which I scheduled interviews agreed to participate, it seems clear that farmers did not feel threatened by the topic or by my association with the environmental management program at Duke.
Research Questions/ Hypotheses
The theoretical framework of adoption-diffusion theory informed my research questions. These questions focused on four general areas: the farmer’s conservation adoption history, his motivations to adopt each conservation practice currently implemented on the farm, his perceived limitations to undertake more conservation, and a self-assessment of conservation successes and needs. Prior to asking these questions, I obtained background information from each farmer pertaining to his farming operation, what crops he was growing, how many acres he was operating, his total sales, and what conservation practices he used (both through participation in government conservation programs and without these incentives).
Data Analysis
My data consisted of transcripts from recorded interviews and field notes from background interviews conducted during the summer of 2006 (Henry and Lurie 2006). Case studies were compiled from the interview transcripts to represent each individual farmer’s conservation history, adoption motivations, perceptions of limitations to conservation adoption, and self-assessment of conservation accomplishments and needs.
A first level of coding was done on the case studies to categorize conservation practices into general categories. Dependent variables from adoption-diffusion literature were then used to code the case studies for the purpose of pattern-matching across cases. Motivations and limitations of adoption were coded based on the farmers’ perceptions of relative advantage, compatibility, complexity, trialability, and observability. Cross-case comparisons were then done for each perception variable to show similarities and differences between the case studies.
Since the sample was small, statistical analysis was not conducted using farmer typologies as independent variables. However, the farmer typologies served as descriptive context for each case. My analysis focuses on illustrating the range of variables influencing farmer decision-making, rather than trying to narrow the focus to a set of predictor variables.
Geospatial Analysis
Farmers drew polygons of conservation impacts and conservation needs on a satellite map of their farm. As previously mentioned, this allowed farmers to talk about actual examples of conservation that they had undertaken and to express limitations that were preventing them from undertaking new practices. Satellite pictures from 2005 were purchased (TerraServer 2007), printed as 2 by 2 foot images and laminated. Farmers were presented with a copy to retain separate from the researcher’s on which responses were drawn. All comments that farmers made while drawing on the maps were incorporated into the qualitative analysis as additional evidence of motivations and limitations.
The polygons showing successes and limitations were then digitized in a geographic information system (GIS). A vulnerability map of the watershed was created using soil data obtained from the Soil Survey Geographic Database (SSURGO), available online from the NRCS (Natural Resources Conservation Service 2007), a 30-meter digital elevation grid obtained from the United States Geological Survey (2007) Seamless Server, and a 2002 cropland data layer obtained from the USDA National Agriculture Statistics Service (United States Department of Agriculture 2002). Only polygons showing impacts and needs related to soil conservation were used for the analysis in order to have a basis for comparison and since this was the concern that farmers most often discussed.
I used the Universal Soil Loss Equation (USLE) as a way to capture vulnerability. This equation uses six variables to gauge soil loss, incorporating soil type and structure, rainfall, slope and slope length, crop type, and conservation practices. I then compared the mean values of the USLE (measured in tons per acre per year of soil loss) on the entire farm compared to the portion where the farmer indicated having improved soil structure from conservation practices. Finally, I quantified the amount of soil erosion reduced by the use of no-till, the practice that farmers indicated as having the most erosion-reduction impacts, to illustrate how much soil erosion would likely be reduced from this type of conservation practice across the watershed (table of variables available upon request).
The fact that relative advantage and observability are the two dominant categories influencing adoption would suggest that policymakers should focus on increasing awareness of relative advantages of innovations and changing visual perceptions of what good stewardship looks like. Programs should use visual aids to influence farmer perceptions of good land stewardship. Nassauer (1992) suggests the use of signs to identify conservation projects and to create a new image of what “good stewardship” looks like. Another strategy is to provide information about conservation practices along with visual images of what these practices can do for the farmer. If policymakers are to increase relative advantage, they need to understand how biophysical aspects of the farm operation are the mechanism through which farmers perceive benefits. If soil quality and soil conservation are two primary motivators, policymakers should also understand that farmers experiencing less erosion would have less incentive for adopting conservation practices that provide these advantages; they should be encouraged to take part in programs that are more appropriate to their landscapes. In short, conservation programs should be framed in a way that shows what benefits farmers gain from undertaking conservation, mostly in the biophysical realm but also in cost and labor savings, rather than in terms of environmental improvements only. Focus should also be on improving the relevancy of regional programs to the landscape – both structurally and biophysically.
Policymakers should also focus on reducing the incompatibility of conservation practices with farmer values and needs. Education and outreach are essential to alerting farmers to how conservation practices can be easily integrated into their farms. Furthermore, district conservationists play a key role in hands-on assistance and relationship-building with the farming community. As program bureaucracy increases, as it has over the last several decades, NRCS staff are becoming more tied to desk work. This will have a serious detrimental affect on the promotion of new conservation practices. Programs should remain as simple as possible, while still meeting conservation objectives, and NRCS staff should continue to serve a hands-on role in promoting awareness of conservation programs.
It is also valuable to consider the stewardship ethic highlighted in these case studies and corroborated by Chouinard (2006). These stewardship motivations are being hampered by the lack of future stability. Since many conservation innovations have delayed returns, it is essential for farmers to feel that it is worth sacrificing profits because of their long-term role as land stewards. Landowner-farmer relationships are fueling this problem, since most farmers rent most of their land and long-term leases are becoming increasingly difficult to secure. Conservation and economic considerations are taking place at two different time scales – conservation requires a long-term vision while current instability in agriculture is requiring farmers to function in a high-risk, short-term mindset. Conservation programs should not only promote good conservation, but should provide long-term stability to farmers.
Incentive programs have been shown to serve a primary role in increasing the rate of adoption of conservation practices. The Conservation Security Program hasn’t required recipients to make many substantial changes to their operations, but it seems to be influencing non-recipients to adopt. The program would be more effective if it found ways to encourage recipients to make more substantial enhancements and had greater expectations for “good actors” to “advertise” their conservation successes to other farmers through visual displays on their farms or through outreach to other farmers.
In conclusion, many farmers are already anticipating a change toward more conservation focus in the Farm Bill and it is causing them to shift toward more conservation-oriented practices. It will be the challenge of future Farm Bills to capture this opportunity to support farmers in a way that promotes long-term stability in agriculture, while achieving environmental objectives.