Final report for GNC20-307
Rancher Preferences for Conservation Programs in Nebraska’s Grasslands
Grasslands are considered one of the most widespread vegetation types in the world (Falconer et al., 1998), and one of the most altered and least protected (Wright, 2013). The conversion of grasslands has many undesirable consequences. Chiefly, soil erosion, the impairment of water resources, and loss of grassland dependent species (Wang, 2017). There is also a heavy emotional toll exacted on communities as rural economies attempt to adapt and the quality of life begins to change.
A range of strategies have been used to influence land use in grasslands. Direct payments to agricultural practitioners for the conservation of landscapes are a longstanding policy instrument (Layton & Siikamaki, 2009). While there are numerous programs for “the conservation of goods and services” across grasslands (Havstad et. al, 2007), evidence shows that the rate of conversion in remaining temperate grassland is occurring five times faster than what can be protected (Lipsey et al., 2015).
Two plausible reasons are that these programs have not appraised all the benefits and services grasslands provide, and they do not sufficiently understand the preferences of those who provide them. Our study aims to bridge this gap by measuring rancher preferences for conservation programming in Nebraska’s grasslands by using an emerging policy program known as Payment for Ecosystem Services (PES). PES is defined as "a transparent system for the additional provision of environmental services through conditional payments to voluntary providers" (Tacconi 2012, p. 35). PES has gained popularity as it seeks to create economic linkages for ranching viability and maximum public and ecological benefit.
This study uses a series of choice experiments, conducted through in-person surveys, to discern what elements of a hypothetical PES program would lead to adoption by ranchers. Our analysis will measure specific attributes (e.g. management practices, contract length, and payment levels) and their perceived applicability to a rancher’s operation.
We believe this study meets all three SARE outcomes through three objectives:
- The analysis of data will provide insights that could reduce inefficiencies in current conservation programming, increasing participation and the profitability of Great Plains ranchers.
- The study partners with ranchers, University researchers, and NGOs to test practical programs that are shown to improve ranch productivity and the resiliency of grasslands.
- The resulting research will demonstrate that management of grasslands and the vitality of rural economies are both a nested public and private interest.
We believe the outcomes of this research will increase learning, awareness, and attitudes toward grassland conservation programs, land-use management, and the development of ecosystem services markets in rural communities. This includes how programs are designed, implemented, and administered by ranchers in the North Central region of the US. This research project addresses the following:
- The preferences measured by the study will increase the knowledge of administrators when designing programs, as the models we test were developed in direct consultation with the ranching and conservation community.
- The associated, voluntary management actions will increase the attitudes and learning of ranchers as they financially benefit through the program’s implementation, and they witness increased resiliency of the landscape over time.
- The potential for behavior change also exists as these models seek to increase participation at a landscape level, regardless of operational capacity, total lands managed, or current financial resources.
- This research will provide foundational information for future conservation program development in grasslands. Furthermore, the resulting information can improve working relationships across the ranching, conservation, and scientific communities.
- (Educator and Researcher)
- (Educator and Researcher)
This study conducted an ex-ante conjoint choice experiment to measure Nebraska ranchers’ preferences for attributes commonly found in payment for ecosystem programs (PES). For the purposes of this research, “ranchers” refer to cattle producers who own or rent native grassland in the state. Our research was informed by the work of Hansen et al. (2018) who focused on PES with ranchers in southeastern Wyoming. Ranchers for this study were recruited through a variety of traditional and nontraditional methods due to the COVID-19 pandemic and health safety protocols. Accordingly, the survey was only available to participants through an online survey platform.
Limitations to in-person gatherings increased our reliance on stakeholder-based organizations such as the Nebraska Cattlemen Association to assist with rancher recruitment for the study. Based on their membership information, we sent invitations to participate in the survey through email, text, and postcards to ranchers operating cow/calf pairs in the state (n=1,548) during two outreach attempts. Additionally, we procured a mailing list from a third-party vendor to send postcard invitations directly to ranchers. Parameters of ranchers engaged in cow/calf operations, on improved pasture or grasslands, of at least 50 acres or more (n=5,743) were selected. These individuals received two postcards (one invitation, one reminder) requesting their participation in the study. We also disseminated invitations for participation through social media, electronic newsletters, and media outlets. With the outreach information that was quantifiable, we determined our study sample to represent 7,291 ranchers in Nebraska.
Limits to in-person gatherings led us to forego all attempts at face-to-face data collection. Our survey was made available only through an online survey platform (Qualtrics) for a duration of 8-weeks. Our objective was to reach ranchers operating in Nebraska, with various levels of expertise, and different operational capacities. However, determining rancher typology from the limited information we could gather was a challenge. Therefore, we found it appropriate to examine known demographic information about ranchers in Nebraska based on the USDA’s (2019) Census of Agriculture, using the North American Industry Classification System (NAICS) standard. This system is the federal statistical standard by which agencies such as the USDA classify a variety of industries, including many of those involved in agriculture.
Using techniques outlined in conjoint choice research, we quantified choice set data to assess marketplace behavior regarding rancher acceptance of PES program attributes. Conjoint choice experiments (CCE), related to conjoint analysis according to Yong (2004), can be traced to random utility theory, discrete choice analysis, and choice modeling. CCE differs from conjoint analysis in that it directly elicits respondent preferences in an effort to better understand the complexities of how products are valued (Louviere & Woodworth, 1983). CCE is sometimes referred to as a discrete choice experiment, but there are noteworthy differences described in the literature. A closer review of these techniques can be found in Louviere et al. (2010).
Attributes in this research refer to the variables that encompass the structural makeup of each hypothetical PES choice set. Study participants were tasked with evaluating a series of choice sets and selecting the most preferred alternative. This allowed us to quantify the perceived value of each attribute with the resulting data informing its “utility” (Yong, 2004). Utility refers to an attribute’s relative worth, which can be expressed numerically. Low values indicate low utility and high values represent increased utility. These utilities are useful for conducting preference simulations, revealing an overall preference share for a PES attribute among study participants. We find this to be one of the primary advantages of CCE as the utility reflects “trade-offs” participants must make when formulating decisions that will affect their ranching operations (Yong, 2004).
The variables that were tested in the hypothetical PES programs were based on the structure of those found in USDA conservation programs. Offerings such as the Conservation Stewardship Program are contractual, encompassing a management activity, compensatory payment, and a length of time for which the arrangement is valid (US GAO, 2007). The contract length and payment level attributes used were based on the model study and our feasibility analysis with ranchers, natural resource professionals, and farm policy experts. The management attributes we tested were derived from recommendations put forth in the state’s wildlife management plan: the Nebraska Natural Legacy Project (Schneider et al., 2011).
Contract length (years)
5, 10, 20
Payment levels (dollars/acre)
5, 25, 50, 100, 250, 500
16 total (see Table 4)
Note. Contract variables for length (3) and year (6) were adapted from offerings currently found in Farm Bill conservation programs. The variables (16) tested for the management attribute were adapted from the Nebraska Natural Legacy Project (Schneider et al., 2011). This resulted in a total of 288 possible choice combinations.
The resulting attributes and variables (Table 1) in aggregate consisted of 288 choice combinations that were possible (16 management; 6 payment levels; 3 contract lengths). Because of this, we utilized an incomplete confounded factorial design to arrive at the choice sets for testing. A common challenge with CCE, one that we address through incomplete confounded factorial methodology, is it asks participants to rate the entirety of attributes present. In this case, reviewing all 288 program profiles would result in survey fatigue and the emergence of bias in a rancher’s selection (Yong et al., 2010). An incomplete factorial design allows us to narrow the total number of profiles presented to a participant, while still testing the main effects of management, payment length, contract duration, and lower order interactions among the three. This technique allows us to elicit responses that are free from subject effect (Kanmongne & Eskridge, 2013).
Survey respondents were assigned to review the choice sets in one factorial array (i.e., block) as a side-by-side comparison of two program profiles. This study examined 128 choice combinations that were represented across eight factorial blocks. These eight blocks were constructed to ensure an even distribution of data across the experiment, allowing us the ability to examine how the three attributes affected participant choices. Within each block, subjects were presented with eight choice sets in which to select their preferred program offering. If a subject did not prefer either program or was unsure about their intentions regarding a choice set, they could select neither.
Participants were also asked a brief series of demographic questions (Table 2) to obtain baseline information about respondents and relative locations in the state. For our purposes, ranchers of all experience levels and operational classifications were of interest. However, one screening question was included to allow those not involved in cattle ranching to self-select out of the survey. As previously discussed, incomplete confounded factorial designs are not conducive to a full estimation of treatment interactions and effects. However, as our interest was narrowly focused on the significance of payment levels, contract lengths, and management actions on preferences, we did not examine correlations of these demographic variables on the programmatic variables we selected for testing.
Which best describes your grazing operation?
Which zip code(s) are most of these lands located?
How many years have you raised and managed grazing animals?
How would you describe the acres your animals graze on?
How many acres are involved in your grazing activities?
Note. To address survey fatigue and anticipated low response rates, this study was narrow in the demographic information collected. These questions were developed through our feasibility analysis within the ranching and natural resources communities.
A noteworthy difference between this research and the model study was in the management attributes that were tested. Hansen et al. (2018) feasibility analysis drew upon direct consultation with various stakeholders in a targeted study area. For this research, which sought to establish a baseline for management across a much larger geographic region (the state of Nebraska), we adopted management strategies that already received considerable stakeholder review from the Nebraska Natural Legacy Project (NNLP). Of additional relevance was NNLP’s emphasis on accentuating biodiversity, a key indicator on the delivery of ecosystem services.
The statistical analysis that was employed in this study is based on McFadden’s (1974) utility model of consumer choice. This was modified to include the random block effects where each respondent is randomly assigned to one and only one block (Kanmongne & Eskridge, 2013). Since the choices are multinomial (choice 1 or 2 or neither), a multinomial logit link function with the base as the "neither" category (C) will be used as the response of which the predictor was a mixed linear model with the factors as the fixed effects and the blocks as the random effects.
The logit model is a generalized linear mixed model and was fit with SAS Proc Glimmix to test for main effects and first-order interactions of the attributes. See Kanmogne and Eskridge (2013) for more details on the statistical analyses of confounded factorial conjoint choice experiments.
Using NAICS (North American Industry Classification System) estimates referenced in the USDA’s (2019) Census of Agriculture, our sample population constituted 63% of the 11,551 ranches engaged in cattle production on grasslands in Nebraska. Over the 8-week survey period, we registered 251 completed surveys from producers—a 3.5% response rate of our study’s sample population. Our geographic assessment (Figure 3) shows respondents were primarily operating in north-central Nebraska (Sandhills region), one of the largest contiguous grasslands in the Great Plains.
In comparison to NAICS data about Nebraska’s ranchers, we found our sample population to be reflective of many of the core demographic characteristics outlined. For example, across categories such as ownership, total acres grazed, type of grazing, and average experience, we find the responses in our sample to mirror statewide trends. For example, 52% of participants reported owning grazing lands utilized in their operation with 77% having more than 20 years of experience. NAICS survey data indicates that 54% of ranchers reported owning the lands they grazed and 74% had 11+ years of experience. In comparing the type of lands grazed, we find similar trends with survey 78% of respondents reporting they grazed rangeland compared to 84% of those in the NAICS survey.
Ranchers self-identified across many of the demographic classifications that were presented (Table 5). A majority of ranchers in this study reported having 20+ years of experience (77%) while conducting their ranching activities on native grasslands (78%). There was a mix in ownership type with 52% reporting they owned the land they grazed, and 33% reporting they utilized a combination of ownership and rentals to manage their herds. No participants utilized grazing allotments on federal or state land, mirroring the larger trend of private land ownership in Nebraska. Additionally, producers represented many different operational capacities. Approximately 48% of ranchers reported they utilized 1,000 acres or more and 88% of all respondents were involved in activities on >100 acres.
Our experimental model detected an indifference for many of the attributes and variables that were tested. Management attributes, however, were found to have a highly significant effect (p<0.0001) on respondents’ preferences for a given program (Table 6). This was observed in Type III tests of fixed effects where preference share of management, payment, length, and the combination of the latter two were examined. In an analysis of payment levels and contract lengths, neither had a significant effect on the respondent’s preference for the choice sets that were presented. This indicates there was a level of indifference to every individual payment or contract attribute regarding participation in our grassland payment for ecosystem services program.
In the broader context of the 16 management practices tested, 61% of participants preferred management actions that were tied to practices known to improve biological diversity such as reduced stocking rate, rotational grazing, stockpiling (Table 7). Conversely, the least preferred practices were related to the management of water resources on the ranch. Approximately 8% of respondents suggested they were willing to remove structures that restricted water movement or remove species or vegetation that had been introduced. Further, participants did not appear to have strong interests in reducing nutrient or insecticide applications to improve water quality (21.97%).
In an examination of lower-order interactions among contract length and payment level, we did not detect significance among these attribute classes or the variables that were tested (Table 8). Additionally, no specific themes about program configurations were apparent (i.e., low payment, short contract length). A leading trend among ranchers in this study was the use of high payments and longer lengths, however, these preferences were not statistically significant among the ranchers we surveyed.
Ranchers must account for a growing number of influences if they are to remain successful in the livestock industry. Grassland conservation programs are no different. Government policy, commodity markets, information networks, technology advancements, and other factors may all contribute to a rancher’s decision to convert native grasslands to alternative uses. However, our study suggests that the significance of financial incentives used to influence those decisions may be overemphasized.
Few in the livestock or natural resources sectors would argue that compensatory payments are not enticing features to offer ranchers engaged in conservation. However, what we found in the literature and confirmed in our study was its importance may be overstated. In the attributes we tested–management, contract length, and payment level–only management variables were found to have significant effects on a rancher’s preference for any given payment for ecosystem services (PES) program. This finding was similarly evidenced in the model study by Hansen et al. (2018), where ranchers self-reported that the nature of the management action and its intended outcomes were of higher importance than the payment level.
Analysis of the relationships among contract lengths and payment levels also yielded no conclusive evidence of the importance to ranchers who participated. Together, these findings suggest the need to offer conservation programs that reflect the challenges of cattle ranching in a natural grassland system. This might also indicate, as the literature confirms, that creating participant flexibility in any program offering is an important design consideration.
This research also provides important clues in how conservation programs can be developed and discussed with ranchers moving forward. For example, using biodiversity both as a metric and a tool for private land conservation. What we discovered is that of all 16 management actions we tested, using grazing to conserve biodiversity was preferred over all other possible program options. This suggests not only that biodiversity is a recognized term within Nebraska’s ranching community, but it also indicates ranchers see biodiversity as a management strategy that can coexist with the core business of livestock production.
In a closer examination of the least preferred management actions, we found the lowest to all to be related to the management of water. This is problematic given water’s importance in agriculture, conservation, and society. Because our work shows that biodiversity is becoming more familiar conceptually within the ranching community, a conservation program offering like PES might look for more ways to specifically address biodiversity needs in relation to aquatic resource management. For example, rather than consultations on individual impaired species, it may be more effective to highlight the overall net decline of the aquatic ecosystem and what these declines tell us about water quality and landscape health.
It is important to note that there will also be trade-offs to consider in relation to ecosystem service production. For example, raising livestock as a provisioning service may at times conflict with others like pest regulation or pollinator services. To be successful, detection and monitoring must be rigorous to ensure the supply of ecosystem services remains constant and is not prioritized for the delivery of any single service. Not only do these programs need to achieve the measurable conservation and delivery of ecosystem services, but these programs need to enhance profitability and demonstrate congruency with ranching lifestyles.
Our review of the literature also sheds light on the need for PES and other conservation programs to create more intentional feedback loops with ranchers. Local participation and decision-making ability are particularly attractive to ranchers, but notably absent in many of today’s conservation offerings (Donlan, 2015). A robust PES program would incorporate stakeholder involvement at several levels (i.e., program design, price negotiations, satisfaction surveys, etc.). This approach mirrors principles laid out in human-centered design and persistence frameworks, ensuring those charged with stewarding natural resources are also protected with technology transfer, financial investment, and localized decision-making ability. With these elements present in the correct proportion, programmatic satisfaction will remain high and bring about the best possibility of behavioral persistence if the option to participate in PES markets remains constant.
The success of a PES program will in part be based on the ability to first meet the needs of livestock producers, then align with other conservation outcomes, and conclude the arrangement with the eventual sale of the ecosystem service (Hansen et al., 2018). In any phase, these programs will see additional benefits by emphasizing ranchers' roles as ecosystem stewards, educating the public and other landowners about the services and marketing opportunities that healthy grasslands provide (Gutwein & Goldstein, 2013). In some cases, PES programs will need to align with pre-existing arrangements or even be started from scratch (Engel et al., 2008). In every situation, programmatic offerings must possess an understanding of local dynamics and support a rancher’s freedom of choice in conjunction with their quality of life (Sorice & Donlan, 2015).
Access entire thesis document:Right Side Up - Payment for Ecosystem Services in Privately Owned Grasslands
Educational & Outreach Activities
We conducted extensive outreach to make this research project successful. This required nine consultations with different stakeholder groups (i.e., ranchers; conservation professionals, nonprofits/advocacy, University researchers) to ensure we could reach a representative sample of the study population. As part of this outreach, we developed three fact sheets for different audiences to help explain the study and its benefits to the livestock and conservation industry. To assist in study participant recruitment, we developed a series of press releases that were disseminated to media outlets statewide as well as through the University’s media and information-sharing network. Our analysis showed seven unique stories were published and shared across media properties. We gave one live radio and one pre-recorded interview for television broadcast.
We have also begun sharing the results of our research with stakeholders in Nebraska. Our principal investigator will defend his findings in two weeks for the completion of his Master of Sciences degree. He is also currently developing a white paper to share results with stakeholder groups; the resulting document will form the basis for a later submission into a yet undetermined research journal.
This project provided a litmus test on the current state of conservation programs in Nebraska. Despite misleading information from elected officials about the intent of conservation programs, we found that ranchers still showed a strong interest in the programs that were modeled. This validates that conservation programs are still relevant and remain popular with today’s ranchers. It also suggests that there is an appetite to provide new offerings that counteract increasingly turbulent economic and environmental conditions.
This project addresses the future of agricultural sustainability in several ways. First, it directly sought input from ranching households. While there are frequent attempts to collect feedback about conservation practices, few efforts attempt to bring ranchers in the fold for the formulation of landscape management strategies. By doing so, we hope to demonstrate one of the pitfalls of modern programs is not using participant focused design which leads to enhanced outcomes for agricultural producers and the lands they manage.
Secondly, this project sought to improve the economic options available to ranchers without the added expense to native landscapes. By using a PES framework, we sought to explore what a new generation of conservation programs could look like. In lieu of targeting single outputs for management, PES emphasizes a broader fencerow to fencerow strategy to improve the health and function of entire ecosystems. This has real implications for future agricultural sustainability as it incentivizes management of entire landscapes alongside the production of the specified commodity.
Lastly, we believe this project addresses agricultural sustainability as it looks at holistic approaches to address local, regional, and even national economic and environmental challenges. If a PES program were effectively designed in Nebraska, it could help boost things like water quality and availability, provide non-traditional revenue for ranchers and increase the tax base, and solidify ranching as a practical and rewarding lifestyle. By seeking out input from ranchers directly, we hope that these findings are then used to formulate the next and future iterations of conservation program offerings.
This project provided new insights on how to effectively design private lands conservation programs in grassland agricultural landscapes. The widely-held opinion, that monetary incentives alone drive conservation program participation, was not confirmed in this project. Instead, we found that the vast majority of ranchers in this study (n=251) chose to participate in a conservation program because of the design of management practices and not because of financial or other contract features.
This is an important finding for our team as monetary payments to farmers and ranchers are often the gold standard for which the viability of conservation programs are assessed. However, as Farm Bill enrollment targets and funding levels decrease through legislative action, relying on financial payments alone is unlikely to generate long-term conservation outcomes for working landscapes or the ecosystems on which they depend.
This finding led us to recommend several emerging frameworks (human-focused program design, behavioral persistence, payment for ecosystem services) that better account for the needs of ranchers while generating long-term conservation outcomes. It also suggests that efforts in sustainable agriculture that overlook the needs of program participants while relying on financial payments to increase participation are likely to produce short-term benefits.
We believe there are an array of contributions that others can make to further the research of PES programs in Nebraska. First, we find it logical to continue research in this field with practitioners who operate working ranches and are currently engaged in conservation stewardship. Building on studies surrounding ranchers’ perceptions of biodiversity and innovation, an effort to create the sociological framework that moves a PES concept to a human-designed conservation program is a critical need.
Another area of need surrounds the creation of the collaborative trust networks that are inherent to successful PES programs. We believe this can be accomplished through securing research funding, which is part of a leveraged approach to assist ranchers with developing pilot locations statewide. These sites, similar to other sites used for technology transfer, should be equipped to serve as the first information exchanges among potential program participants. These sites need to be accessible, replicable and bring together the cadre of entities that will be needed to make PES successful.
The third and largest need will come from the understanding of how to establish the market-based instruments that will lead to PES adoption. This research will need to cover areas of ecosystem service delivery, monitoring, and compliance, as well as how to market these services in a manner that resembles a commodities market exchange. Furthermore, because healthy grasslands exemplify diversity and complexity, it would only make sense for future research to embody an interdisciplinary approach that bridges natural-world capital with land manager motivations.