Consumer Support for Hemp By-Products as Food and Feed

ABSTRACT

Due to changes in legislation, interest in hemp cultivation in the United States (U.S.) has grown in the past five years. Hemp is typically grown for fiber, seeds, or oil, creating significant amounts of cellulosic by-products. To ensure economic viability and environmental sustainability of hemp production, additional uses for these by-products should be identified. Hemp has potential as livestock feed; however, there is a lack of literature on consumers’ perceptions of this practice and on their willingness to purchase the resulting animal food products. The objective of our study was to evaluate U.S. consumers’ perceptions and acceptance of food products from livestock fed hemp. We distributed a survey to U.S. adults (n = 386). Our data demonstrate that consumers overwhelmingly supported growing and feeding hemp to livestock in the U.S. with < 8% opposing either practice. Consumers perceived growing hemp in the U.S. may have economic and environmental advantages, which influences their willingness to purchase food products from animals fed hemp. Most consumers were not familiar with legislation and guidelines surrounding hemp production and many were concerned about exposure to hemp compounds (i.e. cannabinoids) through food. Further research should investigate outreach needed to educate consumers about hemp, especially in the food chain.

摘要

由于立法的变化，在过去五年中，美国对大麻种植的兴趣有所增长. 大麻通常用于生产纤维、种子或油，产生大量的纤维素副产品. 为了确保大麻生产的经济可行性和环境可持续性，应确定这些副产品的额外用途. 大麻具有作为牲畜饲料的潜力; 然而，缺乏关于消费者对这种做法的看法以及他们购买由此产生的动物食品的意愿的文献. 我们研究的目的是评估美国消费者对牲畜饲养的大麻食品的认知和接受程度. 我们向美国成年人（n = 386）进行了一项调查. 我们的数据表明，在美国，绝大多数消费者支持种植大麻并将其喂给牲畜，只有不到8%的人反对这两种做法. 消费者认为在美国种植大麻可能具有经济和环境优势，这影响了他们购买以动物为食的大麻食品的意愿. 大多数消费者不熟悉有关大麻生产的立法和指南，许多人担心通过食物接触大麻化合物（即大麻素）. 进一步的研究应该调查对消费者进行大麻教育所需的外联活动，尤其是在食品链中.

KEYWORDS:

• Livestock
• consumer behavior
• regulations
• hemp by-products
• cannabinoids
• willingness to purchase
• United States

关键词:

• 牲畜; 消费者行为
• 法规
• 大麻副产品
• 大麻素
• 购买意愿
• 美国

Introduction

Cannabis sativa L., commonly known as hemp, is a multipurpose crop that has been cultivated for centuries (Adesina et al. 2020), to produce fiber for textiles and ropes (Campiglia et al. 2020; Leonard et al. 2020) and, more recently, to produce oil and cannabinoids for therapeutic and other uses. Due to the psychoactive components of certain cannabis cultivars and misinformation about the tetrahydrocannabinol (THC) content of industrial hemp, the United States (U.S.) government restricted cannabis production, including fiber-type cultivars, through the 1937 Marijuana Tax Act and 1970 Controlled Substances Act. More recent legislation (i.e., the 2014 and 2018 Farm Bills) defined hemp as an agricultural commodity and removed it from the controlled substances list, developing a pathway for commercialization of hemp production (Tancig et al. 2021). As a result, there is growing interest in cultivating hemp in the U.S (Ajayi and Samuel-Foo 2021).

Industrial hemp is typically grown as a fiber, seed, or oil crop for the textile and allied industries, creating by-products (e.g., leaves, fodder, residual plant fibers). While researchers strive to improve hemp fiber processing, uses for the associated by-products must also be identified to support industry development to an economically consequential level. Reintroducing large-scale hemp processing in the U.S. is anticipated to produce significant amounts of hemp, accompanied by the associated by-products.

Hemp by-products have a high cellulose content and could serve as livestock feed (Kleinhenz, Magnin, Ensley, et al. 2020), potentially enhancing the economic viability and environmental sustainability (Clark 2023; Ely and Fike 2022) of both hemp and food production. Livestock consuming hemp absorb and retain cannabinoid residues in their tissues for months (Kleinhenz, Magnin, Lin, et al. 2020; Krebs et al. 2021). These cannabinoids have been found in the urine of children consuming milk from buffaloes grazing Cannabis sativa L (Ahmad and Ahmad 1990), indicating the ability for cannabinoids to be passed to humans consuming food products from livestock fed hemp.

While there is research on consumer perceptions of hemp products for companion animals (Kogan, Hellyer, and Robinson 2016), there is a lack of literature on how consumers view hemp as a livestock feed. There is a belief among U.S. consumers that hemp and marijuana produce the same products and have the same THC levels and/or psychoactive effects when consumed (Rampold et al. 2021; Stevenson 2017). Further, consumers are confused about the associations between cannabidiol (CBD), THC, and hemp products (Metcalf et al. 2021). Therefore, consumers generally lack basic knowledge about hemp but still associate it positively (Borkowska and Bialkowska 2019).

The quality cues consumers use to purchase meat could impact the success of novel feeds, such as hemp by-products, in entering the animal feeding sector (Aboah and Lees 2020). Preconceived notions about hemp and factors that shape food purchasing decisions suggest that consumers may be fearful of novel hemp foods (i.e., food neophobic) and have feelings about hemp entering the food chain as livestock feed (Metcalf, Wiener, and Saliba 2021). Therefore, the objective of this study was to evaluate U.S. consumers’ perceptions and acceptance of food products from animals fed hemp.

Materials and methods

This study was approved by the [redacted] Institutional Review Board (#8451). We developed an electronic questionnaire-based survey instrument (Qualtrics, Seattle) that used a quantitative approach to data collection. The population of interest was adult (18+ years of age) consumers who live in the U.S. Informed consent was provided prior to completing the survey and identifiable information was not collected. If respondents were not at least 18 years of age and/or did not live in the U.S., responses were not recorded.

Using the current U.S. adult population of 258.3 million (Ogunwole et al. 2021), 385 responses were needed to achieve a 95% confidence level. Ultimately, 404 responses were received, exceeding the calculated sample size, thus not requiring additional procedures to correct for non-response error (Lindner, Murphy, and Briers 2001). After removing responses that were not at least 88% complete (n = 18) and/or nonsensical (n = 1), our final sample size was 386. Prior to distribution, instrument reliability was established through a pilot study using adult U.S. consumers (n = 20).

Pilot study data were analyzed using Statistical Package for the Social Sciences (SPSS; IBM, Chicago) 26.0 software. Cronbach’s alpha reliability coefficient was calculated for two groupings of multi-option questions asking respondents to agree or disagree with prompts following the statements: “I would purchase animal products from livestock that were fed hemp if I knew … ” and “I would purchase products that contained small amounts of CBD or other psychoactive compounds if I knew … .” Reliability coefficients were $\mathit{\alpha }$ = 0.86 and 0.84, respectively, which were classified as good (George and Mallery 2003).

The survey was a researcher-developed instrument that included four sections. The first section of the survey asked prequalifying questions to eliminate bots and identify current dietary practices and shopping habits. Participants were then provided the following explanation of the 2018 Farm Bill from testimony of the Food and Drug Administration (FDA) given before the U.S. Senate (FDA 2019) and asked to agree or disagree with true statements about the 2018 Farm Bill to measure their level of knowledge:

In December of 2018, the 2018 Farm Bill was signed into law. It removed hemp, defined as cannabis (Cannabis sativa L.) and derivatives of cannabis with extremely low concentrations of the psychoactive compound delta-9-tetrahydrocannabinol (THC) (no more than 0.3 percent THC on a dry weight basis), from the definition of marijuana in the Controlled Substances Act (CSA). This law legalized the growing of hemp by American farmers and the USDA (United States Department of Agriculture) regulates and licenses farmers to produce hemp and supports these farmers with access to all the usual programs like crop insurance, farm loans and conservation programs.

The second section of the survey introduced the concept of feeding hemp by-products to livestock then used measures adapted from Fukuda et al. (2023) to assess willingness to purchase and perceived risks and benefits of the products. The third section introduced the concept of CBD and delta-8 before asking additional willingness to purchase questions, revised to focus on these products instead of hemp-fed animal food products. The final section measured demographics of the sample.

The survey was distributed through a paid sampling platform (Prolific, London, England). Demographic filters (i.e., age, ethnicity, gender) were implemented to achieve a representative sample of the national population. The survey was available to respondents on 2023 April 28 from 1400 h to 2120 h until the desired sample size was achieved and respondents received nominal monetary compensation.

Data analysis

Data were analyzed with SPSS 26.0 (IBM Corp., Armonk, NY, U.S.) using descriptive statistics and measures of central tendency. To determine consumers’ willingness to purchase animal food products from livestock fed hemp, exploratory factor analysis (EFA) using the Principal Components Analysis (PCA) method of extraction with Varimax rotation was conducted. Factor analysis models the covariation among a set of observed variables as a function of one or more latent constructs (Bandalos and Finney 2018). Eigenvalues, a measure of variance in the data, correspond to the directions (principal components) that capture more variance in the data and are used to identify where the data varies the most, as represented by the highest eigenvalues (Erdem, Ben Oumlil, and Tuncalp 1999). Reliability of variables constructed using the results of EFA were measured using Cronbach’s alpha, which measures internal consistency with the most reliable measures approaching 1 and those < 0.7 considered unreliable (Hurley et al. 1997; Mackison, Wrieden, and Anderson 2010).

Characteristics associated with consumers’ attitudes about hemp cultivation and as livestock feed were assessed with chi-square analysis in crosstabs for categorical variables or with one-way analysis of variance (ANOVA) for non-categorical variables. Post-hoc tests utilizing Bonferroni or Tukey’s b adjustment were conducted when statistical significance (p ≤ .05) was detected.

Results and discussion

Understanding new markets for hemp and hemp by-products is important to boost availability for food and textile manufacturing. The goal of this study was to provide clarity for a specific use of hemp that is valuable to the entire hemp supply chain – i.e., feeding hemp by-products to livestock. By-products are ubiquitous in livestock systems, increasing the environmental and economic sustainability of food production by upcycling nutrients that would otherwise compete in the human food sector (Drewery 2012) or be landfilled (Ominski et al. 2021).

To address the study goal and objectives, we surveyed a sample population where demographics (Table 1) matched those of U.S. adults for gender and income (Semega and Kollar 2022; Vespa, Armstrong, and Medina 2020). Although we strove for an ethnically representative sample. there was an underrepresentation of Hispanics and Latinos relative to incidence in the U.S. population.

Table 1. Characteristics of sample of consumers in the United States, n = 386.

Gender	Female	48.2%
	Male	48.4%
	Nonbinary	2.6%
	Prefer not to say	0.5%
	Other	0.3%
Age, years	18–24	9.6%
	25–34	20.5%
	35–44	18.4%
	45–54	16.1%
	55–64	20.7%
	65+	14.8%
College student	No	90.9%
	Yes	9.1%
Ethnicity	American Indian	0.3%
	Asian	5.4%
	Black or African American	12.7%
	From multiple races	0.8%
	Hispanic or Latino	6.5%
	White or Caucasian	72.8%
	Prefer not to say	1.0%
	Other	0.5%
Annual household income	<$25,000	13.7%
	$25,000–$50,000	29.0%
	$50,000–$100,000	34.5%
	$100,000–$200,000	16.3%
	>$200,000	2.3%
	Prefer not to say	4.1%
Highest degree earned	Less than high school	1.0%
	High school degree	11.1%
	Some college but no degree	28.2%
	Associate’s degree	10.4%
	Bachelor’s degree	33.7%
	Master’s degree	13.0%
	Doctoral degree	2.6%

Consumer knowledge about food production systems impacts purchasing decisions (Wunderlich, Gatto, and Smoller 2018). To understand respondents’ knowledge about hemp, a short description of the 2018 Farm Bill was provided then respondents were asked more specific questions about the bill. Questions employed a Likert scale with lower numerical means indicating strong disagreement and higher means indicating strong agreement (Table 2).

Table 2. Consumers’ (n = 386) knowledge about the United States (U.S.) 2018 farm Bill.

	Response^a
I am familiar with hemp as a food product.	2.92 ± 1.45
Hemp seeds do not naturally contain cannabidiol (CBD) which is a cannabinoid found in other parts of cannabis plants.	3.36 ± 1.06
The U.S. Domestic Hemp Production Program requires farmers who grow hemp to submit samples for testing to ensure hemp does not exceed acceptable THC levels.	3.65 ± .86
Hemp seed products can be legally used as animal feed, including for livestock.	3.66 ± .95
Hemp produced in the U.S. under the U.S. Domestic Hemp Production Program may be certified as organic under USDA organic regulations.	3.66 ± .80
To produce hemp, you must be licensed or authorized under a state hemp, tribal hemp, or the USDA hemp program.	3.76 ± .93
Hemp seed products can be legally used in the U.S. food supply.	3.89 ± .95
I have not heard anything about the 2018 Farm Bill.	3.99 ± 1.35
The Food and Drug Administration has authority to regulate foods, supplements, drugs, and cosmetics that contain hemp seeds and oils.	4.13 ± .85

^aA 5-point Likert scale was employed such that 1=Strongly disagree, 3=Neutral, and 5=Strongly agree; data are displayed Mean ± Standard deviation.

Respondents indicated they had not heard about the 2018 Farm Bill and were generally unfamiliar with hemp as a food product and with the CBD content of hemp seeds. When asked “I have not heard anything about the 2018 Farm Bill,” the mean response was 3.99 (±1.35), indicating agreement and, when asked “I am familiar with hemp as a food product,” the mean response was 2.92 (±1.45), indicating neutrality. Responses about the U.S. Domestic Hemp Production Program demonstrated participants had vague familiarity as means were within the “Neutral” (3.0) to “Agree” (4.0) categories. More specifically, respondents were asked if they thought farmers were required to submit samples for THC testing (3.65 ± 0.86) and if hemp could be certified organic under the United States Department of Agriculture (3.66 ± 0.80). Respondents were also vaguely familiar with the legality of hemp seed products entering the feed and food chains in the U.S. (3.89 ± 0.95). Of all questions asked, respondents had the most knowledge about the role of the FDA in regulating foods, supplements, and other products that contain hemp seeds and oils (4.13 ± 0.85).

Our data demonstrate that consumers lack knowledge about hemp as a plant and of details of the 2018 U.S. Farm Bill. This agrees with Wysota et al. (2022) who examined young U.S. adults’ knowledge about CBD products, including the 2018 Farm Bill and role of FDA in regulating hemp food products, and revealed misperceptions across the sample population. Further, the Natural Products Association (2019) surveyed U.S. voters and 51% were unaware that the FDA had not approved standards for CBD products, while 41% believed FDA evaluated CBD products and determined safe levels of consumption. While FDA is authorized to regulate hemp products, they have not yet done so to any great extent.

The effect of prior knowledge on accepting hemp as food and feed is not known. In other research, the perceived safety of organic production systems predicted purchasing intent of organic foods (Wunderlich, Gatto, and Smoller 2018). Knowledge or perception of safety may also impact acceptance of hemp entry into the food and feed chains as cannabinoids are retained in animal tissues (Kleinhenz, Magnin, Lin, et al. 2020) and could be passed to humans consuming animal food products. However, there was not a relationship between consumer’s prior knowledge and acceptance of a novel livestock feed in other research (Fukuda, Omana Sudhakaran, and Drewery 2023), which could also be true for hemp. More research investigating how knowledge impacts acceptance of hemp as food and feed should be conducted.

As consumer attitude toward food systems also influences purchasing decisions (Vecchione, Feldman, and Wunderlich 2015), we evaluated if consumers felt that hemp should be cultivated and/or fed to livestock in the U.S. Most study participants (69.4%, n = 268) responded “Yes” to the question “Should hemp be fed to livestock in the U.S.?,” 27.7% were “Neutral” (n = 107), and 2.8% responded “No” (n = 11). Respondents who did not think hemp should be grown in the U.S. (2.8%, n = 11) were asked a follow-up “Select all that apply” question clarifying reasons for their response. Of this sub-group, 89% selected “health reasons,” 55% selected “ethical reasons,” 11% selected “environmental reasons,” and 0% chose “nutritional reasons.” Respondents were also asked if hemp should be fed to livestock in the U.S.; 56.5% of participants (n = 218) responded “Yes,” 36.0% responded “Neutral” (n = 139), and 7.5% responded “No” (n = 29).

Overall, these data indicate that consumers support production of hemp and feeding hemp to livestock in the U.S. While there was more support for the general growth of hemp, rather than as feed, the vast majority of respondents agreed with both practices.

Respondents who responded affirmatively (69.4%) or were neutral (27.7%) to hemp being produced in the U.S. (n = 375) were asked their perceived risks and advantages of feeding hemp to livestock with a “Select all that apply” question (Figure 1). In previous research, perceived risks and advantages significantly impacted attitudes toward genetically modified food (Prati, Pietrantoni, and Zani 2012) and, therefore, may provide insight into consumers’ general behavioral intentions to consume products from livestock fed hemp.

Figure 1. The perceived risks and advantages of feeding hemp to livestock by consumers who were supportive or neutral to hemp being grown in the United States (n = 357).

Perceived advantages outweighed perceived risks − 55.7% (n = 209) participants selected “None of the above” for risks, while 6.7% (n = 25) selected the same for perceived advantages. The most commonly perceived risks of feeding hemp to livestock centered around animal welfare and the quality of animal food products for consumers. Respondents indicated that a risk might be negative effects on the “taste, smell, or quality” (22.4%) or the “nutritional value or healthiness” (15.5%) of the animal food products. Some respondents (21.3%) also felt that hemp may be “unhealthy or unpleasant for the livestock.”

Consumer concern for livestock welfare is growing, and consumers are willing to pay a premium for food products from animals raised under good welfare conditions (Alonso, Gonzalez-Montana, and Lomillos 2020; Bozzo et al. 2019). This may explain why more participants were concerned with the healthfulness of hemp for livestock (21.3%) rather than economic aspects of hemp as feed. Only 10.7% indicated a potential risk of “more expensive animal products” and 11.5% of “reducing availability of hemp for paper or textile production.” Few respondents indicated that a risk of feeding hemp to livestock may be “creating new human diseases” (6.7%).

The most common perceived advantages of feeding hemp to livestock were grounded in environmental and economic benefits. More specifically, 72.0% of participants felt using hemp as livestock feed could be a “helpful way to recycle waste,” 46.7% felt it was a more “natural way to feed livestock,” and 39.5% felt it could reduce the “environmental footprint associated with livestock production.” These data align with previous literature which demonstrates that perceived environmental impact of food production is becoming increasingly important to consumers when making purchasing decisions (Wunderlich, Gatto, and Smoller 2018; Zagata 2014).

The perceived economic advantages benefited both the consumer and hemp producer; 26.7% of respondents indicated that an advantage of feeding hemp to livestock could be “less expensive animal food products” while 56.5% felt it could provide “another form of support for hemp producers.” Respondents also perceived an advantage could be “sparing other feeds that are fit for human consumption” (43.2%) and “nutritional benefits for the livestock” (23.7%).

To understand consumer willingness to purchase food products from livestock fed hemp, respondents were provided the prompt “I would purchase animal products from livestock fed hemp if I knew … ” as a “Select all that apply.” One option (i.e., “I do not support hemp as livestock feed and there are no factors that would positively influence my acceptance of it”) was negatively worded and reverse coded prior to analysis. Items were analyzed through EFA using the PCA method of extraction with Varimax rotation.

Seven items, including “ … that eating hemp didn’t harm the animals” and “ … it would spare other feeds that are fit for human consumption to be used as human food” were cross loaded on more than one factor and were therefore dropped from the analysis. Three factors had eigenvalues > 1, indicating these items had the highest contrast when split into three factors. Variables made with items in each of the three factors were reliable with Cronbach’s alphas > 0.70, which is considered acceptable (Table 3). Three separate measures were then created with the remaining eleven items out of the original 19 that had a factor loading > 0.5 and were not cross loaded onto another factor.

Table 3. Exploratory factor analysis of consumers’ (n=386) willingness to purchase animal products from livestock fed hemp in the United States (U.S.).

Variable	Item	Mean	SD	Factor loading
	I would purchase animal products from livestock fed hemp if I knew …
Supporting Agricultural Systems	… the hemp supported the natural balance of our eco-system	3.94	1.04	0.75
(Cronbach’s alpha = 0.90, variance explained = 47.54%, Eigenvalue = 9.04)	… livestock producers wanted to feed hemp to their animals	3.67	1.04	0.75
	… the hemp was certified organic	3.67	1.07	0.74
	… livestock enjoyed eating the hemp	3.99	1.04	0.72
	… it would create new jobs for the U.S.	3.89	1.03	0.69
Consumer Experience
(Cronbach’s alpha = 0.81, variance explained = 9.61%, Eigenvalue = 1.83)	… the taste, feel, smell, appearance, and quality of the animal product were the same	4.11	1.00	0.71
	… it would lower the cost of animal products (e.g., beef, milk) for consumers	4.10	1.04	0.69
Regulatory Trust
(Cronbach’s alpha = 0.87, variance explained = 5.89%, Eigenvalue = 1.12)	… regulatory officials certified the hemp as not containing any CBD	3.51	1.07	0.84
	… regulatory officials certified the hemp as not containing any THC	3.68	1.06	0.79
	… animal products produced from animals fed hemp were specifically labeled	3.51	1.07	0.70
	… hemp was a small percentage of the animal’s diet	3.35	0.94	0.61
	… the hemp was healthy for the livestock	4.29	0.96	Cross loaded
	… I do not support hemp as livestock feed and there are no factors that would positively influence my acceptance of it^a	4.28	1.12	Dropped
	… it would lessen the environmental impact of livestock production	4.09	1.05	Cross loaded
	… it would spare feeds that are fit for human consumption to be used as human food	3.82	1.03	Cross loaded
	… a brand I trusted endorsed or carried products from livestock fed hemp	3.62	1.03	Cross loaded
	… I was confident that eating hemp didn’t harm the animals	4.27	0.98	Cross loaded
	… leading scientists and dietitians endorsed it	3.92	1.05	Cross loaded
	… I heard more about it through social media and other popular press outlets	2.84	1.12	Cross loaded

^aItem was reverse coded before inclusion in analysis.

The first variable influencing willingness to purchase was that feeding hemp to livestock would provide or strengthen local jobs and the desire to use hemp-byproducts; this variable was called Supporting Agricultural Systems and included five items such as “ … it would create new jobs for the U.S.” and “ … livestock producers wanted to feed hemp to their animals.” The item “ … hemp supported the natural balance of our eco-system” was similar to a cross-loaded item “ … it would lessen the environmental impact of livestock production.” The difference may be the positive framing, the support that hemp provides to agricultural producers. This suggests that the variable is related to positive feelings about agriculture.

The second variable, called Consumer Experience, described the impact of hemp as livestock feed on the consumer experience of resulting food products, including the cost of products and sensory experience. This variable originally included three items but “ … I do not support hemp as livestock feed and there are no factors that would positively influence my acceptance of it” was dropped after reliability analysis revealed that it lowered Cronbach’s alpha from 0.81 to 0.75. The remaining two items “ … the taste, feel, smell, appearance, and quality of the animal product were the same” and “ … it would lower the cost of animal products for consumers” are related to the experience consumers would have with the food products from livestock fed hemp.

The final variable, called Regulatory Trust, was related to consumers being unwilling to be exposed to animal food products that contain cannabinoids without their knowledge and was formed from four items related to the impact of hemp on the resulting animal food products. Two items directly express that willingness to purchase would be influenced by regulatory officials certifying the hemp as not containing CBD or THC. Labeling of products (i.e., “ … animal products from animals fed hemp were specifically labeled”) and inclusion of hemp in the animal’s diet (i.e., “ … hemp was a small percentage of the animal’s diet”) also seemed to be related to consumers unwillingness to be exposed to hemp in the food chain unknowingly or in large quantities.

The emergence of the Regulatory Trust variable was interesting because U.S. consumers tend to distrust government regulations of food safety and may be skeptical of programming from governmental organizations (Costanigro et al. 2014). However, a sample of consumers who were informed that the FDA has not regulated CBD or approved standards for production reported they were now 44% less likely to use CBD products (Natural Products Association 2019). Ellison (2020) reported that 71% of surveyed licensed hemp producers ranked “Regulatory issues” as an extremely important area of research. Cumulatively, our data and existing literature suggest that U.S. consumers lack knowledge about the regulations surrounding hemp production but, along with hemp stakeholders, feel this is an important area of focus.

The variable Regulatory Trust was chosen for further analysis because, while previous studies of consumer interest in sustainable fiber products have considered both consumer experience (Sigaard and Laitala 2023) and support of agriculture (Bernard, Hustvedt, and Carroll 2013), specific issues related to hemp have not been examined. The four items within this variable were summed and rescaled to create a single measure, Regulatory Trust. The mean for the Regulatory Trust variable was 3.51 (±0.85) which indicated that respondents were neutral in how regulatory or labeling concerns impacted their willingness to purchase animal products from livestock fed hemp. Participants were split at the median (3.33) into three groups with participants below the median (n = 153) labeled “Low Trust” (2.66 ± 0.52), those at the median (n = 48) labeled “Median Trust” (3.33 ± 0), and those above the median (n = 185) labeled “High Trust” (4.24 ± 0.45).

ANOVA was conducted using the three Regulatory Trust groups to explore how respondents with different levels of trust responded to various questions such as support of hemp production in the U.S. or familiarity with the 2018 Farm Bill (Table 4). Tukey’s b post-hoc testing did not identify significant differences between the three groups in levels of knowledge of the 2018 Farm Bill. The three Regulatory Trust groups were also not significantly different in their dietary habits, openness to trying new foods, food opinion leadership, or bonus food spending. These variables were adapted from Fukuda et al. (2023) except for bonus food spending which was from Horowitz and McConnell (2003). Most importantly to this study, support of hemp as livestock feed was not significantly different between participants across levels of Regulatory Trust.

Table 4. Effect of regulatory trust group on consumers’ (n = 386) willingness to purchase animal products from livestock fed hemp in the United States^a.

Item	Low trust (n = 153)		Median trust (n = 48)		High trust (n = 185)
I would purchase animal products from livestock fed hemp if I knew …	Mean	SD	Mean	SD	Mean	SD
… the hemp supported the natural balance of our eco-system	3.71^a	1.19	3.77^a	0.91	4.17^b	0.89
… livestock producers wanted to feed hemp to their animals	3.41^a	1.12	3.46^a	0.94	3.94^b	0.92
… the hemp was certified organic	3.38^a	1.15	3.44^a	1.01	3.97^b	0.92
… livestock enjoyed eating the hemp	3.82^a	1.19	3.88^a	0.96	4.16^a	0.89
… it would create new jobs for the U.S.	3.64^a	1.11	3.71^a	0.94	4.15^b	0.91
… the taste, feel, smell, appearance, and quality of the animal product were the same	3.78^a	1.17	4.06^a	0.91	4.41^b	0.76
… it would lower the cost of animal products (e.g., beef, milk) for consumers	3.80^a	1.22	4.04^a	0.97	4.37^b	0.08
… regulatory officials certified the hemp as not containing any CBD	2.64^a	0.69	3.00^b	0.58	4.37^c	0.67
… regulatory officials certified the hemp as not containing any THC	2.71^a	0.72	3.63^b	0.61	4.50^c	0.58
… animal products from animals fed hemp were specifically labeled	2.94^a	1.02	3.54^b	0.80	3.96^c	0.95
… hemp was a small percentage of the animal’s diet	2.74^a	0.82	3.38^b	0.64	3.85^c	0.85
… the hemp was healthy for the livestock	4.03^a	1.17	4.23^ab	0.63	4.52^b	0.76
… I do not support hemp as livestock feed and there are no factors that would positively influence my acceptance of it	1.76^a	1.16	1.58^a	0.92	1.70^a	1.01
… it would lessen the environmental impact of livestock production	3.87^a	1.26	3.92^a	0.85	4.31^b	0.86
… it would spare other feeds that are fit for human consumption to be used as human food	3.51^a	1.17	3.6 7^a	0.88	4.12^b	0.85
… a brand I trusted endorsed or carried products from livestock fed hemp	3.16^a	1.04	3.48^b	0.92	4.03^c	0.88
… I were confident that eating hemp didn’t harm the animals	3.97^a	1.21	4.35^b	0.70	4.50^b	0.74
… leading scientists and dietitians endorsed it	3.54^a	1.15	3.73^a	0.84	4.28^b	0.86
… I heard more about it through social media and other popular press outlets	2.55^a	1.07	2.71^a	1.13	3.11^b	1.13

^aGroups sharing letters across a row are statistically identical based on a Tukey’s b post-hoc test after ANOVA.

Level of Regulatory Trust was related to participants’ willingness to purchase animal products from livestock fed hemp. In only two cases were there no differences between the three groups: the outright unwillingness to purchase these animal products and that the “… livestock enjoyed eating the hemp”. In half of the items suggested to the participants as potentially influencing their willingness to purchase, the “Low Trust” and “Median Trust” groups were statistically the same, and the “High Trust” group had a significantly higher willingness to purchase. These included both items included in the Consumer Experience variable and all of the items in the Supporting Agricultural Systems variable except for enjoyment, which was not statistically different across Regulatory Trust groups.

Given that the Regulatory Trust variable was used to create the split, it is not surprising that the three groups were significantly different from each other across all four items. The item that had the highest level of reported influence on willingness to purchase across all three groups was that “…the hemp was healthy for the livestock,” suggesting that the lack of difference in animal enjoyment was not a lack of interest but perhaps related to concern for animal welfare.

We explored characteristics that were related to respondents’ attitudes about hemp cultivation and use as livestock feed (Table 5). Respondents were asked, “Should hemp be grown in the U.S.?” and “Should hemp be fed to livestock in the U.S.?” Male and female respondents were included in the analysis with other gender identities excluded due to low incidence. Gender was not significantly associated with attitude toward hemp cultivation in the U.S., but there was a significant difference (p ≤.01) for the response to “Should hemp be fed to livestock in the U.S.?” such that a greater proportion of males felt that hemp should be used as livestock feed, while females tended to be neutral. This contrasts findings that gender did not impact acceptance of insects, a novel feed, as livestock feed (Fukuda, Omana Sudhakaran, and Drewery 2023) but agrees with research indicating that males were more open to genetically modified food than females (Moerbeek and Casimir 2005).

Table 5. Characteristics that significantly affected consumers’ (n = 386) attitudes toward hemp cultivation or hemp use as livestock feed in the United States (U.S.)^a.

		Should hemp be grown in the U.S.?			Should hemp be fed to livestock in the U.S.?
		No (n = 11)	Neutral (n = 107)	Yes (n = 264)	No (n = 29)	Neutral (n = 136)	Yes (n = 217)
Gender^b		—- p = .39 —-			—- p = .01 —-
Male	n = 187	2.7%	25.1%	72.2%	4.8%^a	31.6%^a	63.6%^a
Female	n = 186	3.2%	31.2%	65.6%	10.8%^a	39.2%^b	50.0%^b
Ethnicity^c		—- p = .01 —-			—- p = .09 —-
Minority	n = 101	3.0%^a	46.5%^a	50.5%^a	9.9%	42.6%	47.5%
White or Caucasian	n = 281	2.8%^a	21.4%^b	75.8%^b	6.8%	33.1%	60.1%
Annual household income		—- p = .05 —-			—- p = .70 —-
<$25,000	n = 53	0.0% ^a	32.1% ^b	67.9%^a	7.5%	41.5%	50.9%
$25,000–$50,000	n = 112	5.4% ^a	20.5% ^a	74.1%^a	8.9%	30.4%	60.7%
$50,000–$100,000	n = 133	3.0% ^a	27.8% ^b	69.2%^a	7.5%	33.1%	59.4%
$100,000–$200,000	n = 63	0.0% ^a	41.3% ^b	58.7%^a	6.3%	46.0%	47.6%
>$200,000	n = 9	11.1%^b	11.1% ^a	77.8%^a	11.1%	33.3%	55.6%
Prefer not to say	n = 16	0.0%^a	18.8% ^a	81.3%^a	0.0%	43.8%	56.3%

^aCharacteristics only displayed if they significantly (p ≤ .05) affected consumer attitude; groups sharing a letter within column are statistically identical based on a Tukey’s b post-hoc test after ANOVA.

^bThirteen responses excluded due to low incidence for other gender categories.

^cFour responses excluded due to low incidence of “Prefer not to answer;” ethnicities except for “White or Caucasian” were combined to develop the variable “Minority.”

When considering whether or not the respondent was the primary food buyer in their household, there were no associations with attitude toward hemp cultivation (p = .87) or feeding hemp to livestock (p = .98).

To analyze the effect of ethnicity on respondents’ attitudes, we combined all ethnicities except “White or Caucasian” into a single variable called “Minority” as described by Drewery et al. (2023). There was a significant relationship (p ≤.01) between ethnicity and attitude toward hemp cultivation such that a greater proportion of minority respondents were neutral and a greater proportion of White or Caucasian respondents affirmed that hemp should be grown in the U.S. There was not a relationship between ethnicity and attitude toward hemp as livestock feed. A study conducted on Canadian and U.S. consumers indicated that White populations were 1.12 times more likely to have used CBD products in the past year than all other ethnicities combined (Goodman et al. 2022), perhaps explaining our findings of greater agreement with hemp cultivation from White/Caucasian, rather than minority, populations. To the best of our knowledge, there is not yet research examining the role of ethno-racial identity on perceptions of hemp products.

There was a significant association (p = .05) between annual household income and attitude about hemp cultivation in the U.S., but not for attitude about hemp being used as livestock feed (p = .70). A significantly greater proportion of respondents with an annual household income of ≥$200,000 did not think hemp should be grown in the U.S. as compared to other income groups. However, this income group also had the smallest proportion of “Neutral” respondents and highest proportion who affirmed “Yes,” that hemp should be grown in the U.S. This indicates that high-income consumers were decisive and polarized in their support of hemp cultivation in the U.S. Respondents with household incomes of <$25,000 or $100,000–$200,000 were more likely to be neutral to hemp cultivation.

Kolodinsky and Lacasse (2021) reported a significant positive association between incomes of $50,000–$75,000 or $75,000-$100,000 with support of hemp, although all income groups were more supportive of hemp than the lowest group (<$25,000). However, in other research, there was not an association between income and acceptance of hemp foods (Lacasse 2021). Our findings of a relationship between income and support for hemp production may be related to the perception that hemp products are more expensive and the ability of higher income brackets to afford said products (Kim and Mark 2018). However, Kolodinsky and Lacasse (2021) hypothesized that consumers with higher incomes can avoid living near hemp fields that may emit odors or have other drug-related concerns, thus potentially driving their greater acceptance over lower-income populations.

Openness to trying new foods was also significantly associated with attitude toward hemp cultivation in the U.S. (p = .02), but not attitude about hemp as livestock feed (p = .17). There were parallel increases for openness and for acceptance of hemp cultivation – i.e., greater proportions of respondents who were open to consuming new foods also believed that hemp should be grown in the U.S. There was not a relationship between age (p ≥ .22), being a college student (p ≥ .42), highest degree attained (p ≥ .19), or region (p ≥ .98) and either attitude towards hemp cultivation or towards hemp as livestock feed. There was also no association between attitude (p ≥ .12) and response to “How often do family or friends ask for your opinions about food products?.”

Our study builds on the scant repository of literature that focuses on consumer perception of hemp entering the food and feed chains. However, the study is limited in the reliance on a convenience sample that may not be generalizable to the U.S. population. Further research should be conducted on a larger scale, perhaps by a private marketing firm, to provide more insight into consumer perceptions of hemp entering the feed and food chains.

Conclusion

These findings indicate that consumers support both the cultivation of hemp and feeding hemp to livestock in the U.S. The economic and environmental benefits of integrating hemp into food systems seem to be important to consumers; they are guiding factors in the willingness to purchase products from hemp fed livestock. However, concern about exposure to cannabinoids in the food chain will need to be squarely addressed to increase acceptance. Consumers lack knowledge about hemp as a plant and of the details of the 2018 U.S. Farm Bill which includes regulatory controls intended to prevent cannabinoid exposure. Much of the marketing for hemp products is designed to promote products that can legally skirt prohibitions that apply to marijuana in certain regions. However, efforts to ensure that hemp production is separate and differentiated from legalized marijuana production are not being broadly publicized. One approach to allay consumer concerns may be to provide hemp education and outreach and explain procedures in place to ensure hemp is safely integrated into the food and fiber supply chain in the U.S.

Authors Contribution

MD and GH contributed equally to this work, jointly receiving funding, developing the experimental design, analyzing data, and synthesizing the manuscript.

Ethical approval

This study was approved by the Texas State University Institutional Review Board (#8451).

Consent

Informed consent was provided prior to completing the survey and identifiable information was not collected.

Highlights

• This research fills a deficit in the literature on consumer perceptions of hemp cultivation and use as a livestock feed in the United States (U.S.), which is of growing interest given recent legislation that relegalizes the growth of hemp in the U.S.

• Findings indicate overwhelming consumer support for hemp growth and use as livestock feed in the U.S. while also revealing perceived risks and advantages associated with these practices.

• Consumers are largely unfamiliar with regulations surrounding hemp in the U.S. and, thus, education efforts are likely important as hemp enters the food and, potentially, feed chain.

Acknowledgments

The authors acknowledge support from Dr. Nicole Wagner.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Texas State University Multidisciplinary Internal Research Grant (MIRG) Program.