1. Introduction
The equine veterinary profession in the United States faces a concerning decline in equine practitioners. Currently, the health of the estimated 6.7 million horses in the U.S. (American Horse Council Foundation, 2024) is managed by 3,972 equine-specific veterinarians, along with 4,667 mixed-species veterinarians (McKay and Grice, Reference McKay and Grice2024). However, more than 50% of American Association of Equine Practitioners members fail to renew their membership within 5 years of graduating from veterinary school (Grice, Reference Grice2020). In addition, the number of new graduates entering equine practice is also declining. In 2023, 1.3% of new veterinary school graduates planned to enter equine practice full-time, compared to 5.7% in 2003 (McKay and Grice, Reference McKay and Grice2024). There are many potential factors contributing to this trend; some may be external, such as a declining horse population, while others are internal, such as non-competitive salaries, high debt-to-income ratio, poor work–life balance, and occupational stressors (Grice, Reference Grice2020).
While this issue is complex, one crucial part of the solution relates to the demand for equine veterinary services and the corresponding financial health of equine veterinary practices. However, little is known about horse owners’ demand for veterinary services and how this demand influences the financial viability of equine practices; most research in the field of veterinary economics is relatively recent and focuses on the companion animal sector. Knippenberg (Reference Knippenberg2014) provides an overview of the relationship between consumer demand for veterinary services and factors such as market size, income level, consumer perceptions, and the prices of complementary and substitute goods.
Understanding demand for equine veterinary services is important for the development of the profession. For example, Wiltzius et al (Reference Wiltzius, Blackwell, Krebsbach, Daugherty, Kreisler, Forsgren and Moyer2018) attributed stagnant demand growth in the veterinary profession in part to consumers’ lack of financial access to veterinary services. In this paper, we examine whether measures of financial capacity are related to willingness to pay. Previous research suggests that there is low utilization of certain financial risk management tools such as equine health insuranceFootnote 1 and colic reimbursement programsFootnote 2 (Stowe, Reference Stowe2021; Stowe, Kibler, and Barrowclough, Reference Stowe, Kibler and Barrowclough2022), as well as a limited prevalence of dedicated equine healthcare savings accounts (Gibson, Adam, and Stowe, Reference Gibson, Adam and Stowe2025).
To our knowledge, no studies have considered the determinants of demand for veterinary services, equine or otherwise. This current study uses survey data to examine horse owners’ WTP for typical equine veterinary services, including vaccinations, lameness exams, and colic surgeries, while considering a wider range of sociodemographic, horse ownership, and financial factors that may influence demand. By better understanding the demand for equine veterinary services, practices can be more strategic in terms of pricing, marketing and advertising, client education, and provision of services, which can ultimately lead to demand growth.
1.1. Background – Equine Veterinary Services
Equine veterinarians provide medical care for horses, ponies, mules, and donkeys; their services include preventative care like administering vaccinations, specialized care such as diagnosing and treating illnesses and injuries, and performing elective or emergency surgeries. Unlike small animal veterinary practice, where pets are seen at the clinic for routine health care, regular equine veterinary health care is often ambulatory; a recent survey showed that nearly 80% of respondents’ veterinarians travel to the horses’ locations for routine health care (Gibson, Adam, and Stowe, Reference Gibson, Adam and Stowe2025). This model has economic implications for both veterinarians, who face travel time and fuel costs, and horse owners, who benefit from convenience but may incur additional farm call fees. Horses are typically transported to equine veterinary clinics only for specialized care, diagnostics, and surgeries, adding another layer of cost and logistical considerations for owners.
According to the 2024 AVMA/AAEP Report on the Economic State of the Equine Veterinary Profession (McKay and Grice, Reference McKay and Grice2024), there were an estimated 3,785 equine veterinarian in the United States, 60% of whom were female. As one might expect, there are higher concentrations of equine veterinarians in parts of the country with dense horse populations. According to the American Horse Council Foundation’s 2023 Economic Impact Study of the U.S. Horse Industry (2024), Texas has the most horses, followed by California and Florida. These three states also had the greatest concentration of equine veterinarians (McKay and Grice, Reference McKay and Grice2024). Moreover, nearly 2/3 of equine veterinarians are solo practitioners; only 4% work for large corporate franchises of veterinary hospitals (McKay and Grice, Reference McKay and Grice2024).
Horse owner demand for equine veterinary services may be determined by a number of factors, including standard sociodemographic characteristics like age, income, education level, and location, as well as horse ownership characteristics, such as the number of horses owned, the age and value of the horses, and the primary purpose of the horses. It may also depend on whether the service itself is considered routine, elective, or urgent in nature. In this paper, we use survey data to examine factors of willingness to pay (WTP) for three equine veterinary services, which span service types as well as cost ranges. Annual vaccinations are used as an example of a routine service, lameness exams represent an elective service, and emergency colic surgery represents an urgent service. In each category, the objective was to identify a service which would be familiar and relevant to most horse owners.
The American Association of Equine Practitioners provides guidelines for annual core and risk-based vaccines for horses. When administered by a veterinarian, the cost of annual vaccinations for a horse ranges from about $70–$200 per horse (Equine Veterinary Fee Survey, 2023); the range in prices can be explained by practice-level variation as well as the number and types of vaccines administered, as recommendations for risk-based vaccines differ according to a number of factors. There are few substitutes for vaccines administered by veterinarians; horse owners can purchase vaccines and self-administer, or they can choose to undervaccinate or not vaccinate at all.Footnote 3
Lameness exams are an example of an elective service. Owners of horses with an unresolved lameness are limited in their ability to ride, train, and compete. Additionally, there are welfare concerns for horses that are uncomfortable even when not being ridden. However, lameness is not typically an urgent life-or-death condition. Lameness exams vary in expense depending on practice-level differences, thoroughness of the physical exam, and follow-up diagnostics such as radiographs, ultrasounds, and MRIs. A rough estimate for a lameness exam with no diagnostics ranges from $50–$400 (Equine Veterinary Fee Survey, 2023). There are few substitutes for a lameness exam; often, the next best alternative is for the horse to be rested to see if the lameness will resolve itself.
Emergency colic surgery is an example of an urgent service. Colic is a general term for pain in the horse’s abdomen; this abdominal pain can be caused by a variety of issues. Most colics are not surgical, but in a small percentage of cases, colic is life-threatening and cannot be resolved without surgery. The cost of colic surgery can vary dramatically depending on the complexity of the surgery itself, as well as peri-operative care and potential post-operative complications. In typical uncomplicated cases, costs might range from $8,000–$15,000 across the U.S. (Adam, Reference Adam2024). If the owner elects to forgo treatment, the only alternative is euthanasia. With the higher cost range, this procedure constitutes a larger portion of the horse owner’s income than the other two services studied in this paper. Favorable outcomes are not guaranteed (Freeman, Reference Freeman2018), and the age and use of the horse may factor into the owner’s decision. Finally, the decision to pursue colic surgery is an urgent and often very emotional decision.
2. Methods
2.1. Survey Tool
An online survey developed in Qualtrics was used to collect data for the study. The target audience was U.S. residents ages 18 and older who were currently financially responsible for at least one horse, pony, mule, or donkey at the time of the survey. After pilot testing with a focus group of 7 horse owners, the final survey was approved by the institution’s Office of Research and Integrity (protocol #88436). The link to the online survey was distributed through participating organizations’ social media and email lists. Participating organizations included the University of Kentucky Ag Equine Programs, American Horse Publications, and the State Horse Council Advisory. The authors also shared the invitation to participate and the survey link on their personal social media pages. The survey was open from August 15, 2023, through September 11, 2023, and garnered 4,992 responses across all 50 states.
The survey tool consisted of five sections designed to gather comprehensive data on horse ownership, veterinary service utilization, and horse owner demographics. The first section, “Horse Ownership,” gathered data on respondents’ years of experience with horses, activities with their horses, where their horses reside, and financial considerations related to equine healthcare. The second section focused on respondents’ usage of equine veterinary services, such as frequency of visits, as well as attitudes towards potential changes in current practices, such as using the veterinarian on call rather than the primary veterinarian or willingness to use telemedicine.
In the “Willingness-to-Pay” section, respondents were presented with a payment card displaying a range of prices for three equine veterinary services: routine vaccinations, elective lameness exams, and emergency colic surgery. The payment card approach is a common elicitation technique for estimating willingness to pay for agricultural goods and services (e.g., Hu et al (Reference Hu, Woods, Bastin, Cox and You2011), Yang et al (Reference Yang, Hu, Mupandawana and Liu2012; Reference Yang, Qing, Hu and Liu2013)). The payment card approach avoids the starting point bias inherent in dichotomous choice approaches, is straightforward to implement in an online survey format, and allows individuals to “shop around” for the value closest to their maximum willingness to pay (Boyle and Bishop (Reference Boyle and Bishop1988), Donaldson, Shackley, and Abdalla (Reference Donaldson, Shackley and Abdalla1997)). Given the wide variation in horse ownership experience within our sample, the structured nature of the payment card was deemed more appropriate for ensuring data quality across all respondents. In addition, the payment card approach is often employed to increase completion rates and reduce protest bias and strategic zeros as compared to open-ended elicitation methods (Arrow et al (Reference Arrow, Solow, Leamer, Portney, Radner and Schuman1993), Donaldson, Thomas, and Torgerson (Reference Donaldson, Thomas and Torgerson1997), Randall, Hoehn, and Tolley (Reference Randall, Hoehn and Tolley1981)). However, it is important to note that when using the payment card approach, results may still be affected by a range bias, a centering bias, and/or an endpoint bias (Mitchell and Carson (Reference Mitchell and Carson1989), Hu, Zhong, and Ding (Reference Hu, Zhong and Ding2006)).
The specific price intervals on each payment card were deliberately tailored to the service in question: vaccinations, lameness exams, and colic surgery. The number of price levels and the monetary increments were based on current market prices and refined through pre-testing. This was done to provide respondents with a realistic set of price anchors for each distinct service, with the primary goal of eliciting the most accurate WTP possible for each scenario. As a result for annual vaccinations, the WTP ranged from $0 to $500 or more in $50 increments. For lameness exams, the price ranged from $0 to $1,000 or more in $100 increments. For emergency colic surgery, a more extensive range was used due to the higher cost of this procedure, which started at $0 and increased in $2,500 increments up to $30,000 or more.
The fourth section collected standard demographic information, including the respondent’s location, age, household income, gender, race/ethnicity, educational attainment, and number of dependents. Finally, a section on “Awareness & Concern” asked respondents to assess their prior knowledge of challenges facing the equine veterinary profession, their current level of concern about the situation, and any recent difficulties they might have experienced in obtaining veterinary care.
2.2. Geographic Data
Due to the ambulatory nature of equine medicine, we include two geographic attributes in the demand analysis. First, we generate an indicator variable for Census Bureau Division, which is determined by respondent’s state of residence. There are nine such divisions across the U.S. (United States Census Bureau, 2024). These regional indicators can account for broad geographic variations in WTP across the US, which may be driven by differences in the cost of providing services or by the prevalence of different equine breeds and/or disciplines. However, they do not capture the important dimension of rurality. This distinction may be important because it has been reported that horse owners in remote areas are disproportionately affected by the decline in equine veterinarians (Vigoroux, Reference Vigoroux2023).
Initially, we measured rurality using Rural Urban Commuting Areas (RUCA) codes (Economic Research Service, 2019). However, this classification assigned about 70% of the sample to the three metropolitan areas. To address this limitation, we generated an index of relative rurality (IRR). The original IRR is composed of county-level population, population density, urbanization, and distance to the nearest metropolitan/micropolitan statistical area; it yields a continuous measure of rurality ranging from 0 to 1, with 0 representing the most urban areas in the dataset and 1 representing the most rural areas (Waldorf, Reference Waldorf2006).
Since our survey collected respondent ZIP codes rather than counties, we followed the approach of Inagami et al (Reference Inagami, Shasha, Hassan, Shendge Martine, Probst Janice and Stone Roslyn2016) in adapting the IRR to the ZIP code level (IRR_ZIP). The IRR_ZIP variable is composed of the population, population density, and distance to the nearest urban center. ZIP code level population and population density were obtained from the 2020 US Census (Ruggles et al, Reference Ruggles, Flood, Sobek, Backman, Chen, Cooper, Richards, Rodgers and Schouweiler2024). Urban areas are identified using US Census Bureau data and were obtained in a TIGER/line shapefile (United States Census Bureau, 2024). Distance to the nearest urban area was computed by measuring the distance from the centroid of the respondent’s zip code to the centroid of the nearest urban area using QGIS. This approach allows for a more precise and granular assessment of rurality, capturing the varying degrees of remoteness and access to urban centers that may influence horse owners’ demand for veterinary services.
2.3. Empirical Model
We adopt a Random Utility Model (RUM) as the theoretical foundation for our analysis (Manski, Reference Manski1977). The RUM posits that an individual’s utility is comprised of a deterministic component (based on observable attributes) and a random, unobservable component. In this context, a horse owner’s willingness to pay for a veterinary service is derived from the utility they receive from ensuring their horse’s health through the targeted services.
Specifically, this framework directly leads to our empirical model, where the WTP is a function of the owner’s socioeconomic characteristics and horse ownership attributes. We employ an interval regression model to estimate factors of horse owners’ WTP for equine veterinary services. The payment card method used in the survey provides WTP data in the form of intervals, where the respondent’s true maximum WTP for each service lies within the selected range. Interval regression is well-suited for this type of data as it accounts for the inherent uncertainty in the true WTP values. Unlike ordinary least squares (OLS) regression, which estimates a single point estimate, interval regression utilizes a maximum likelihood estimator for each interval, providing a more flexible and accurate representation of the data. In addition, the interval regression approach allows for censoring from respondents when they select the highest available option in the payment card, acknowledging the possibility that their true WTP exceeds the upper limit of the provided range.
For the interval regression, the unobserved latent variable is estimated by
$$ Y_{i}^{*}=\beta X+\varepsilon, $$
where ε ∼ N(0,σ 2 I). The probability of selecting each interval is
$$ \Pr \left(lwtp\lt {Y_{i}}^{*}\lt uwtp|X\right). $$
The lower and upper bounds of the WTP interval, lwtp and uwtp, correspond to the veterinary service under analysis.Footnote 4 Since interval regression uses the maximum likelihood estimator, homoscedasticity and normality are assumed. The term X represents the set of explanatory variables, which includes demographic and horse ownership attributes. Respondent demographic attributes consist of income, age, highest education level attained, dependents in the household, Census Bureau Division, and IRR_ZIP. Horse ownership attributes include number of horses owned, years of experience with horses, self-reported market value of a horse owned, whether the horse is insured or enrolled in a colic reimbursement program, financial tolerance for an equine medical emergency, and dedicated equine health savings account. The dependent variable is the respondent’s maximum WTP for service i. WTP models were estimated using Stata IC 15.1 (64-bit).
The log-likelihood function for the interval regression model being maximized is:
$$\ln {\rm L}=\sum _{j\epsilon R}\ln \left\{1-\Phi \left({y_{Rj}-X_{j}\beta \over \sigma }\right)\right\}+\sum _{j\epsilon I}\log \left\{\Phi \left({y_{2j}-X_{j}\beta \over \sigma }\right)-\Phi \left({y_{1j}-X_{j}\beta \over \sigma }\right)\right\},$$
where the first term represents the likelihood for right-censored observations (j ∈ R) and the second represents the likelihood for interval observations (j ∈ I). The y j are the unobserved observations, and Φ( ) is the cumulative standard normal distribution.
With ordered categorical data, an ordered probit model might be considered as an alternative empirical approach. However, this approach is less appropriate and powerful for our payment card data, as it discards valuable information. An ordered probit model would inappropriately treat the WTP intervals as simple ranked categories (e.g., 1st, 2nd, 3rd) and ignore the actual monetary values that define them. In contrast, the interval regression model fully utilizes these known and meaningful dollar amounts as the interval boundaries. As established in the contingent valuation literature, this superior use of information makes interval regression a more statistically efficient estimation method than discrete choice models for payment card data (Alberini, Reference Alberini1995; Cameron and Huppert, Reference Cameron and Huppert1991; Yang et al., Reference Yang, Hu, Mupandawana and Liu2012). Given these methodological advantages, we concluded that interval regression was the most appropriate and robust model for our analysis.
3. Results
Table 1 provides a description of the variables used in the regression model along with the summary statistics. In this table, the sample includes the 3,967 respondents who answered at least one of the WTP questions as well as all of the questions eventually used as explanatory variables.Footnote 5 All variables are categorical with the exception of IRR_ZIP, which has a minimum value of 0.06 and a maximum value of 0.80. The most frequently reported annual household income category was $100,000–$149,999, while the least common was less than $25,000 (24 and 4%, respectively). Age categories 25–34, 35–44, 45–54, and 55–64 generated a fairly uniform rate of responses (18%–22% each), but only 5% of respondents were in the youngest age category of 18–24. The majority of respondents had taken some college courses or earned a college degree (57%), and another 37% had graduate or professional degree coursework. 46% of respondents reported having at least one dependent. The most responses came from the South Atlantic Census Division (22%), and the least came from the New England Census Division (5%).
Variable descriptions and summary statistics (n = 3,967)
About 78% of respondents owned between one and five horses, while 68% reported having at least 20 years of experience with horses. Respondents were asked to select one of their horses to use for the WTP questions. 16% of these horses were insured, and 10% were enrolled in a colic reimbursement program. About 29% of the horses were reported to have a market value less than $5,000, while another 29% had a market value between $10,000 and $25,000. 17% of the identified horses had a market value of $25,000 or more. When asked about the highest amount of medical costs the respondents would finance if this horse experienced a medical emergency today, 42% of respondents indicated $1,000–4,999, followed by 32% at $5,000–$9,999. Finally, about 13% of respondents reported having a dedicated savings account for equine health expenses.
Table 2 presents the distribution of WTP responses for each of the three veterinary services; there are differences in the number of participants responding to each question, so the relevant sample size is noted. For annual vaccines, the most common value selected was $150 (26.5%), followed by $100 (21.4%) and $200 (19.2%). Only 2.1% of the sampled indicated a maximum WTP of $0. For a lameness exam, the most common value selected was $200 (28.5%), followed by $300 (22.5%) and $500 (14.8%). Only 0.3% of the sample indicated a zero WTP. Finally, for colic surgery, the most common value selected was $5,000 (26.2%), followed by $2,500 (18.7%) and $10,000 (16.1%). Nearly 16% of the sample reported a maximum WTP of $0. Over 70% of the sample reported a maximum WTP of $7,500 or less, which is notable because in a best-case scenario, the minimum cost of a colic surgery would be about $8,000.
Distribution of WTP responses
Table 3 presents the results for estimated predictors of WTP for vaccinations (column a), lameness exams (column b), and colic surgeries (column c).Footnote 6 Examining the results collectively suggests that some attributes are common correlates of demand across service types, while other attributes exert service-specific effects.
Determinants of willingness-to-pay for select equine veterinary services
***,**, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.
a Reference category: less than $25,000
b Reference category: 18–24 years old
c Reference category: high school or less
d Reference category: Pacific (AK, CA, HI, OR, WA)
e Reference category: less than $5,000
f Reference category: less than $1,000
3.1. (a) Vaccinations
Many sociodemographic factors appeared to be correlated with WTP for vaccinations. WTP was an average of $18.36 higher for respondents with an income of $150,000–$249,999 (p<0.05) and $23.63 higher for income of at least $250,000 (p<0.05) as compared to respondents with income less than $25,000. In general, age displays an inverse relationship with WTP. As compared to respondents in the youngest cohort (18–24 years), WTP for vaccinations declines across older age groups. Those aged 35–44 were willing to pay an average of $15.51 less (p<0.1), while those aged 45–54, 55–64, and 65 or older exhibited even greater reductions in WTP, averaging $21.67 (p<0.01), $29.42 (p<0.01), and $21.32 (p<0.05) less, respectively. The highest level of education attained by respondents is also correlated with WTP for vaccinations. Respondents with at least some graduate coursework or a graduate degree were willing to pay an average of $16.05 more (p<0.05) than those with a high school level of education, while those with a professional degree were willing to pay $17.99 more (p<0.01). The presence of dependents in the household is associated with a lower WTP, with respondents having at least one dependent willing to pay $8.25 (p<0.01) less than respondents with no dependents.
There were regional and urban/rural differences in WTP for vaccinations at a 10% significance level or better. As compared to respondents in the westernmost states included in the Pacific Census Division, respondents residing in five other regional divisions demonstrated a higher WTP, ranging from $11.15 (East South Central, p<0.1) to $77.05 (New England, p<0.01). Conversely, respondents in the Mountain Division were willing to pay an average of $12.50 less (p<0.1). Rurality, as measured by the index of relative rurality (IRR_ZIP), was also a factor in WTP differences. For each 0.1 unit increase in rurality, WTP decreased by an average of $4.40 (p<0.01). This suggests that the most rural respondents would be willing to pay an average of $44.04 less than the most urban respondents.
Respondent horse ownership attributes also explain some variation in WTP for vaccinations. Respondents who own 1 to 5 horses were willing to pay an average of $28.92 more (p<0.01) than those with more than 5 horses. Interestingly, respondents with at least 20 years of experience with horses were willing to pay $7.49 less (p<0.05) than respondents with less experience. The self-reported value of the respondent’s horse is related to WTP for vaccinations in an unexpected way. Respondents with horses valued in the $10,000–$24,999 range were willing to pay $9.17 less (p<0.05) than respondents with horses valued at less than $5,000, while respondents with horses valued at $25,000 or more were willing to pay $11.40 less (p<0.05).
Finally, factors related to financial access to equine health care demonstrate significant associations with WTP for vaccinations. Respondents with a horse enrolled in a colic reimbursement program were willing to pay $10.67 more (p<0.1) than those who did not utilize those programs. Respondents with a dedicated equine savings account were willing to pay an average of $12.71 more (p<0.01) than those without. Furthermore, WTP is generally increasing across the emergency expense tolerance categories. Compared to those who can withstand an emergency of less than $1,000, those with a tolerance of $1,000–$4,999 were WTP $16.89 more (p<0.01), those with a tolerance of $5,000–$9,999 were WTP $26.80 more (p<0.01), a tolerance of $10,000–$19,999 generated an average WTP of $33.40 more (p<0.01), and those with a tolerance of at least $20,000 were WTP an average of $23.41 more (p<0.05).
3.2. (b) Lameness Exams
Income has a more nuanced influence on WTP for lameness exams than vaccinations, with nearly all levels being significantly different from the base category of less than $25,000. A clear positive trend emerges, with WTP increasing as income rises. Respondents reporting annual household income in the $50,000–$74,999, $75,000–$99,999, and $100,000–$149,999 range were willing to pay $65.14 (p<0.01), $72.34 (p<0.01), and $71.67 (p<0.01) more, respectively. At higher income ranges, the increase in WTP was more pronounced. Respondents reporting income in the $150,000–$249,999 range were willing to pay $101.70 more (p<0.01), while those with incomes of at least $250,000 were willing to pay $93.65 more (p<0.01).
Similar to vaccinations, age demonstrates a consistent negative association with WTP for lameness exams across all age categories. Compared to the 18–24 reference range, WTP is lower for all age groups. For those in the 25–34 age range, the average premium is –$40.58 (p<0.05); the 35–44 age range, –$52.09 (p<0.01); the 45–54 age range, –$48.66 (p<0.01); the 55–64 age range, –$32.77 (p<0.1); and for respondents aged 65+, the average decrease in willingness to pay was $44.94 (p<0.05). Unlike vaccinations, educational attainment does not appear to have a significant relationship with WTP for lameness exams. However, similar to vaccinations, the presence of at least one dependent is associated with a WTP of $18.18 less (p<0.05) than respondents with no dependents.
Once again, there were regional variations in WTP. As compared to the Pacific Division, respondents residing in the New England area were willing to pay an average of $54.51 more (p<0.01), all else equal. Conversely, respondents in five other divisions exhibited lower WTP at the 5% significance level or better, with differences ranging from –$25.63 (South Atlantic, p<0.05) to –$51.35 (West South Central, p<0.01). However, variation in WTP according to the relative rurality of the respondents was not statistically significant.
Among the horse ownership attributes, the number of horses owned remains a significant predictor of WTP. Respondents owning 1–5 horses were willing to pay an average of $32.75 more (p<0.01) than those owning more than 5 horses. Those utilizing a colic reimbursement program were WTP $22.07 more (p<0.1), while those with an established equine savings account were willing to pay an average of $22.66 more (p<0.05) than respondents who did not. Similar to vaccinations, WTP was generally increasing across emergency tolerance categories as compared to the reference category of less than $1,000: the average premium was $46.61 (p<0.01) for those able to withstand emergency expenditures of $1,000–$4,999; $72.69 (p<0.01) for a tolerance range of $5,000–$9,999; $93.48 (p<0.01) for a tolerance range of $10,000–$19,999; and $85.85 (p<0.01) for respondents able to withstand emergency expenses of at least $20,000. Neither years of experience with horses nor horse value significantly influences WTP for lameness exams.
3.3. (c) Emergency Colic Surgery
For emergency colic surgery, the effect of income appears to be less granular compared to the other services, with only the highest income bracket demonstrating a significant difference from those in the reference category of under $25,000. Respondents reporting an annual household income of at least $250,000 were willing to pay an average of $1,123.73 (p<0.01) more.
The age range of the respondents also influenced WTP, but with a less consistent pattern than observed for vaccinations and lameness exams. Compared to the reference age range of 18–24, respondents in the 25–34 range were willing to pay $611.66 less (p<0.1), while those in the 45–54 age range demonstrated an even greater reduction of $1,073.06 (p<0.01). Similarly, those aged 55–64 and 65+ were willing to pay $690.34 (p<0.1) and $1,071.50 less (p<0.01), respectively. Similar to lameness exams, educational attainment did not appear to be statistically related to WTP for emergency colic surgery, although respondents with at least one dependent had an average WTP of $220.89 less, all else equal.
There were regional differences in WTP for emergency colic surgery. As compared to respondents in the Pacific Division, WTP was less for respondents in six other areas at the 5% significance level or better, ranging from –$511.71 (South Atlantic, p<0.05) to –$961.92 (East South Central, p<0.01). The degree of relative rurality was not statistically significant.
Among horse ownership variables, respondents owning 1–5 horses were willing to pay an average of $541.93 more (p<0 .01) for emergency colic surgery than those owning more than 5 horses, all else equal. Those with at least 20 years of experience with horses were willing to pay –$268.80 less compared to those with less than 20 years of horse experience, although this result was only significant at the 10% level.
The value of the respondent’s horse significantly influenced WTP for emergency colic surgery in a predictable fashion, increasing across value categories. Compared to respondents with horses valued at less than $5,000, those with horses valued $5,000–$9,999 were willing to pay an average of $605.39 more (p<0.01), those with horses valued $10,000–$24,999 were willing to pay $663.05 more (p<0.01), and those with horses valued at $25,000 or more were willing to pay $1,191.48 more (p<0.01).
While insurance was found to have no statistically significant effect on WTP for colic surgery, enrollment in a colic reimbursement program was associated with a higher WTP, averaging $404.68 more (p<0.1). Similarly, respondents with a dedicated equine health care savings account were willing to pay an average of $715.67 more (p<0.01).
Finally, the ability to financially withstand an equine medical emergency affected WTP for emergency colic surgery in an expected manner. Compared to respondents with a financial tolerance of less than $1,000, WTP was increasing across tolerance categories. Those with a tolerance of $1,000–$4,999 were willing to pay $1,568.40 more (p<0.01), a tolerance of $5,000–$9,999 yielded an increase in WTP of $4,955.90 (p<0.01), a tolerance of $10,000–$19,999 produced an increase in WTP of $8812 (p<0.01), and a tolerance of at least $20,000 results in an average increase in WTP of $10,926.09 (p<0.01).
4. Discussion and Conclusion
Understanding the correlates of demand for equine veterinary services is important for both practitioners and horse owners. For veterinarians, our results suggest that the demand for their services is significantly correlated with factors like owner income, horse value, and geographic region. Based on the segment of the client base, this could inform tailored communication strategies, the development of tiered wellness packages, or financial counseling that recognizes the diverse economic realities of their clients. For horse owners, the findings highlight the significant role that financial preparedness plays in the ability to afford care. For example, our result that having a dedicated equine savings account is a strong predictor of higher WTP for both elective and emergency services can empower owners by demonstrating how proactive financial planning can directly translate into a greater capacity to provide for their horses’ well-being in times of need.
Across the three models, one of the consistent results is that more robust financial access to equine veterinary services is associated with increased demand, be it through savings, insurance or reimbursement programs, or income. Moreover, of the three “insurance” measures (equine health insurance, colic reimbursement program, and dedicated equine savings account), respondents with a dedicated equine savings account had the greatest demand for equine veterinary services as evidenced by the highest WTP premiums. Combined with annual household income, these resources contribute to a horse owner’s ability to accommodate an unexpected medical expense and are of vital importance in providing access to more expensive treatments and emergency services. Understanding this link between financial access and demand should motivate horse owners to be disciplined in setting aside funds for horse health care; equine veterinary practices can also leverage this finding to educate clients about the importance of financial planning for their horses’ healthcare needs and encourage the use of savings strategies, insurance options, and reimbursement programs to facilitate proactive and comprehensive care.
Interestingly, the age of respondents is also a contributing factor in WTP variance across the three equine veterinary services examined in this study. Although this result is not entirely consistent, in general, WTP for all three services is highest in the youngest age category. This finding may reflect generational differences in attitudes towards veterinary care or varying levels of risk aversion among different age groups.
Another noteworthy result relates to geographical differences in WTP. Actual costs for procedures can vary according to cost of providing these services, amount of competition from other practices, and practice-level variation, among other things; those existing differences in actual costs may have anchored respondents’ reported WTP. It has been suggested that the decline in equine veterinarians has disproportionately affected rural horse owners; however, the degree of rurality only influenced WTP for vaccinations, with respondents in the most rural areas being willing to pay around $45 less per horse than respondents in the most urban areas. It is possible that lower demand for routine services might influence veterinarians’ decisions to serve remote areas for other, less common procedures. Somewhat surprisingly, rurality was not significantly related to WTP for lameness exams or emergency colic surgery. Qualitative data collected in the survey’s comments section provides some preliminary insights, with some respondents indicating that their $0 WTP for emergency colic surgery was due to the impracticality of transporting a horse in distress to a distant surgical facility. This highlights the need to consider access challenges and logistical constraints faced by horse owners in remote areas.
Horse ownership variables were also important determinants of WTP for equine veterinary services. Respondents with fewer horses were willing to pay more for each service; with fewer horses, owners have overall reduced horsekeeping costs and may have more financial capacity to care for each individual horse. Interestingly, respondents with more years of horse experience were generally willing to pay less, although this result was not significant for lameness exams. While the reasons for this result are not immediately clear, it is possible that past negative experiences with certain procedures, such as colic surgery, may have influenced their willingness to pay for those services. On the other hand, respondents with extensive experience may have the knowledge to manage more of the care themselves or may have established long-term relationships with veterinarians that generated service discounts. The value of the horse played a role in WTP, but not always in the expected direction. However, for high-cost, life-saving treatments like colic surgery, owning a higher-valued horse appears to be associated with an increase in WTP. This finding likely reflects the financial rationality of investing in a costly procedure for a horse with a higher market value.
Using the results from the models in Table 3, the expected WTP for the average respondent in our survey can be determined for each service.Footnote 7 This respondent has an average WTP of $205.65 for annual vaccinations, $399.13 for a lameness exam, and $4,433.66 for emergency colic surgery. This latter result deserves a brief discussion, as it illustrates how lack of financial access limits demand growth. Both our result (average WTP = $4,433.66) and that in Fahey (Reference Fahey2018), where Colorado horse owners’ average WTP for colic surgery was estimated at $2,610, illustrate a critical tension. The average horse owner would prefer not to euthanize their horse, but the average WTP is significantly less than the best-case minimum cost of the procedure, which ranges from $8,000 - $15,000 depending on location in the US. In our sample, just under 40% of respondents reported a maximum WTP of $7,500 or more, while only 23% reported a maximum WTP of $10,000 or more. This implies that a substantial proportion of horse owners would face the difficult decision of euthanasia due to financial constraints, even when a potentially life-saving procedure is available.
While this study provides valuable insights into predictors of demand for equine veterinary services, it is important to understand its limitations. First, as in any stated preference approach, there is the potential for hypothetical bias, where respondents tend to overestimate their true WTP (Loomis, Reference Loomis2011). In an attempt to moderate this effect, a statement was included in the instructions of the choice experiment, emphasizing the importance of providing realistic responses, as these may inform future decisions in equine veterinary practices. While statements such as this have been shown to be effective in reducing hypothetical bias for some study participants (Champ, Moore, and Bishop, Reference Champ, Moore and Bishop2009), it remains possible that respondents’ reported WTP may not align with their actual behavior in a real-life scenario. Moreover, while the payment card approach was utilized to avoid the starting point bias inherent in a dichotomous choice approach, it may be affected by a range bias, a centering bias, and/or an endpoint bias (Mitchell and Carson (Reference Mitchell and Carson1989), Hu, Zhong, and Ding (Reference Hu, Zhong and Ding2006)). Additionally, while this survey received strong participation, our ability to determine the representativeness of the sample is limited. There is no data resource detailing the profile of horse owners and horses in the United States like those available for human populations, such as the US Census. Therefore, caution should be exercised in generalizing the findings to the broader population of horse owners. Finally, we focus on three specific equine veterinary services – one routine, one elective, and one urgent. Obviously, veterinarians offer a much larger suite of services, and it would be unreasonable to expect predictors of demand for these three services to be the same across services in the same category. For example, castration of a male horse would be different example of an elective service. Only a small subset of horse owners are involved in breeding, so this service would be relevant only to a select few. In addition, the motivations for castrating a horse are far different from having a lameness exam performed. Taken together, it would not be surprising if factors influencing demand for these two services were dissimilar. However, the approach used in this study can be applied to other services.
Solving the challenges in the equine veterinary industry, including providing more competitive salaries for equine veterinarians, will require a multi-pronged approach. A deeper understanding of demand for services, as provided in this study, is the initial component of this approach, empowering practices to implement strategic pricing, marketing, and client education initiatives that foster revenue growth, stimulate demand, and ultimately support financial stability for equine veterinarians. A valuable direction for future research would be to apply our findings to a market-level analysis, such as by simulating a demand curve to estimate the likely uptake of these veterinary services at different real-world prices, offering direct insights for practitioners’ pricing strategies. Together, by promoting financial awareness and preparedness among horse owners and by tailoring service offerings to meet the diverse needs and constraints of different client segments, the equine veterinary profession can work towards a more sustainable and accessible healthcare system for horses.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/aae.2025.10026.
Acknowledgements
The authors would like to thank Amy Grice, VMD, MBA, Charlotte McKay, and participants in the focus group for their contributions in developing the survey tool and, of the course, the nearly 5,000 U.S. horse owners who took the time to complete the survey.
Author contributions
Conceptualization, O.L.G. and C.J.S.; Methodology, O.L.G., C.J.S., and S.Z.; Formal Analysis, O.L.G. and C.J.S.; Data Curation, O.L.G. and C.J.S.; Writing – Original Draft, O.L.G., C.J.S., and S.Z.; Writing – Review and Editing, C.J.S. and S.Z.; Supervision, C.J.S. and S.Z.; Funding Acquisition, C.J.S.
Financial Support
This work is supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, Hatch Project under 1014277.
Data Availability Statement
Data for the study were obtained through a survey tool developed by the authors. The survey tool is available as a supplementary file. Due to confidentiality restrictions under IRB, the raw data used in this study cannot be made publicly available. However, aggregated data that support the findings of this study are available from the corresponding author upon reasonable request.
Competing Interests
The authors of this article had no personal or financial relationships that could cause a conflict of interest.
AI contributions to research
No AI-assisted technologies were used in the generation of this manuscript.
Open access
