Local or Ecological? An Agri-Food Choice Experiment for Extra Virgin Olive Oil

Abstract

Agricultural production systems and certificates of origin determine consumers’ preferences for agri-food products, attributes which have mainly been investigated separately. In this study, we examined the joint effect of these two quality criteria and how they influence agri-food choices. To this end, a choice experiment exercise was applied for extra virgin olive oil agri-food in a protected geographical area of south-eastern Spain where non-conventional production systems (organic and integrated pest management) and product origin labels (protected designation of origin and sustainable local brand) associated with the territory are being used. The results demonstrated that consumers’ choices to purchase extra virgin olive oil were primarily influenced by the type of production system, with a clear preference evident for organic systems. Furthermore, consumer preferences for sustainable local over generic national certificates of origin were also confirmed. In concrete terms, the willingness to pay for organic extra virgin olive oil is more than double that of the conventional one, while the surcharge for local origin may rise up to 75%. This work contributes to the understanding of how information related to the type of production system and its interaction between different origin certification labels applies in agri-food products, offering broader implications for producers and policy makers.

1. Introduction

There has been considerable change in food consumption trends over the last decade due to an increasing interest among consumers in their dietary choices [1]. For instance, consumers are progressively cognizant of the fact that their dietary decisions can have long-term implications for both their personal health and environmental sustainability [2,3]. The increased demand for ecologically sustainable foods is related to an increased interest in the efficient use of resources, and consequently in the future well-being of the population [4]. Nevertheless, the extent to which consumers value and respond to ecological food products through consistent value-based behavior, especially towards agri-food products [5], remains a matter of some debate [6].

Several factors have been identified as having the capacity to influence consumer behavior in relation to the selection of ecological agri-food products. These include the higher price of such products, which is often perceived as a limiting factor in terms of their purchase [7]. More specifically, one of the differentiating attributes of ecological agri-foods is their production systems [8]. In agricultural production systems, ecological food comes from systems where natural resources are managed in a way that preserves the functions of the agroecosystem to ensure its sustainability over time and where no chemicals are used [4,9]. Even the degree to which ecological agri-food products contribute to environmental sustainability can be calibrated, differentiated between organic and integrated pest management systems.

Another of the quality attributes that can define ecological agri-foods is certificates of origin from a geographical area, something that consumers cannot verify themselves, either before or after purchase [10]. This attribute serves to enhance the credibility of the product. In order to credibly communicate this characteristic, most agri-foods have at least one certified quality label [11], but hardly ever are different certificates of origin investigated and compared. Consequently, consumers can differentiate between food quality on the basis of their belief that the certificate of origin is associated with this quality [12], in a manner analogous to brand or price [13,14]. A substantial body of research [15,16,17] has been conducted on the impact of quality certificates granted to specific geographical areas on consumer dietary preferences. The findings of these studies indicate that consumers exhibit a higher propensity to purchase products bearing geographical certificates when compared to those without such certification.

Despite the existence of an increasing number of choice experiment studies [5,18,19] (which currently represent one of the most popular stated preference methods for food product/attribute valuation) that have explored the effects of consumer behavior on the purchase of ecological agri-foods, few of them have specifically considered the joint influence of both the type of production and the quality certification of a geographical area on such behavior. However, their results are not conclusive at all. For instance, some of them expose stronger preferences and higher willingness to pay for locally produced foods compared to those labeled as organic or environmentally friendly [20,21]. In contrast, other research highlights that while ecological certifications command a premium, attributes like local production and contribution to rural development have less influence [18,22].

In this study, we examined whether the production system and origin labels of food are relevant to agri-food choices. To this end, a choice experiment method was applied for the extra virgin olive oil agri-food in a protected geographical area of south-eastern Spain where non-conventional production systems and product quality labels associated with the territory are being used. Also, the performance of different origin labels is also assessed and compared.

To our knowledge, this is the first study to examine the role of consumer preferences towards the integration of different and complementary geographical certification of agri-food quality together. If we find that there is a direct influence of these attributes on the agri-food purchase decision, this would be an important finding, as it would imply that future studies of choice experiments should also jointly consider the inclusion of these attributes if they are to improve the understanding of consumers’ purchasing behavior towards ecological and environmentally sustainable agri-foods. In this sense, the present study provides an empirical basis for the promotion of strategies for the production and marketing of ecological agri-foods. The promotion of such strategies is intended to be in the interests of sustainable development of the territory, while satisfying social demands for a quality agricultural product.

2. Materials and Methods

2.1. Case Study

Extra virgin olive oil has been selected as one of the most representative agri-foods of the Mediterranean diet, which is consumed by many people worldwide and is a fundamental pillar in the agri-food system of many Mediterranean countries, including Spain [23]. Spain is responsible for 70% of European Union production and 45% of global production, making it an undeniably significant economic contributor [24]. However, beyond its economic implications, the production of Spanish olive oil carries substantial social, environmental and territorial consequences [25].

At the regional level, there are numerous territories that cultivate olives in a sustainable and environmentally responsible manner, thereby offering a premium ecological product that bears the distinctive seal of a particular geographical area [23]. There are different types of quality certification of geographical areas, including protected designation of origin (PDO) certificates, which identify a food product to its geographical origin, and which stipulate that at least one of its stages of production, processing or preparation is carried out in the defined geographical area. This is a very widespread certification in European countries, such as Spain and Italy, and has national recognition.

In this study, the specific case of a protected geographical area in the Region of Murcia (south-eastern Spain) called Sierra Espuña Regional Park is examined, where a local quality certification emerges to promote not only local production and social values but also environmental sustainability and rural development. This is the ‘Territorio Sierra Espuña’ brand, which promotes good practices that favors biodiversity and environmental sustainability and distinguishes the commercialization of quality local products [26]. This territorial brand also supports the development of rural areas and improves the lives of the local population [27]. Thus, this geographical area combines areas where there are more local quality certificates associated with the territory, together with nationally recognized certificates located in the vicinity of the study area.

2.2. Choice Experiment Method

Choice experiments comprise a methodological framework grounded in multi-attribute and random utility theory [28]. Dealing with a choice experiment exercise, consumers are presented with a series of goods or services, which are defined according to a set of attributes, and are asked to select the one that provides them the highest utility level. As such, the utility provided by a good or service could be obtained as the sum of the utility from the attributes integrating such good or service [multi-attribute utility theory]. From these choices, the preferences for the valuated goods or services are disentangled, thereby deriving the utility function from any combination of attributes and the relative importance of each of them [29]. This utility function is not only defined by the attributes themselves (observed variables) but also by individuals’ tastes and circumstances (unobserved variables) [random utility theory].

The present study investigates the influence of attributes associated with agricultural production systems and certificates of geographical area on consumer behavior to extra virgin olive oil. It seeks to calculate the perceived utility of each attribute, in addition to its relative importance and economic valuation. The inclusion of an additional monetary attribute allows us to estimate the willingness to pay for each attribute [30].

2.2.1. Experimental Design

The first stage in the implementation of a choice experiment exercise comprises the definition of attributes and their levels. To achieve that, a comprehensive literature review was first undertaken [31,32,33,34] with the purpose of identifying the most valued characteristics and attributes of the product. This review confirmed the attributes associated with the production system and quality certificates of geographical areas are used individually in the case of the choice of agri-foods. In this sense, 4 attributes were selected for the specific case for the purchase of one liter of extra virgin olive oil in the case study: (i) Agricultural production system, (ii) Protected Designation of Origin, (iii) ‘Territorio Sierra Espuña’ Brand and (iv) Price (EUR/L). Other significant attributes for understanding consumer choices for extra virgin olive oil (e.g., taste, package, branding…) were disregarded because of the purpose of this study—production systems and certifications of origin—which should be recognized as a limitation.

To validate this selection, a focus group was held with local experts who confirmed the suitability of these attributes for the case study. This focus group also reached a consensus on the attribute levels to be used in the choice experiment. Thus, the first of the attributes considered, the production system, was divided into three levels: conventional, integrated and organic. The two attributes regarding geographical area of production were articulated in a dichotomous mode, with two levels related to the presence or absence of each of them in the product. In order to determine the levels of the monetary attribute, the price of extra virgin olive oil of different characteristics in hypermarkets, supermarkets and specialized websites was consulted, taking into consideration prices of 6, 8, 10 and 12 EUR/L.

Table 1. Attributes and levels definition.

Attributes	Levels	Definition
Production system	Conventional	Olives come from conventional agricultural systems, characterized by mechanization, monocropping and the use of synthetic inputs, such as chemical fertilizers or pesticides, in accordance with the levels established by legislation for this type of agriculture.
	Integrated	Olives come from orchards grown under integrated pest management, which make maximum use of natural resources and production mechanisms, ensuring long-term sustainable agriculture. Biological and chemical methods for pest control are carefully chosen and balanced, considering the demands of society, profitability and environmental protection.
	Organic	Olives come from organic farming, where natural resources are managed in a way that preserves ecosystem functions to ensure their sustainability over time. No chemicals are used.
Protected Designation of Origin (PDO)		This sign identifies those products whose quality and characteristics are related to the geographical area where they are produced. Their production, processing and preparation processes are controlled according to regional standards. These certificates tend to have a national projection.
‘Territorio Sierra Espuña’ (TSE) Brand		This is a local label that promotes sustainable tourism, and the commercialization of quality local products from the surroundings of Sierra Espuña Regional Park. It ensures the adoption of good practices in favor of biodiversity and sustainability, supporting the development of rural areas and improving the standard of living of the local population.
Price	6, 8, 10, 12 EUR/L	Market price for a liter of extra virgin olive oil packaged in a glass bottle

lists the attributes and levels used in this work. This information was also provided to respondents before choice experiment responses.

The selected attributes and levels were combined using a Bayesian D-efficient design, which seeks to minimize the standard errors of the parameters of a fixed effects choice model. To apply this design, the results of a pilot survey choice experiment were used to obtain a first approximation of the parameters. The implementation of the experimental design was conducted utilizing Ngene software v.1.2.1 [35].

The choice experiment was designed with 12 choice sets, which were grouped into 2 blocks of 6 choice sets each, randomly distributed during data collection. Each choice set was composed of three generic alternatives, of which two represented different types of extra virgin olive oil, and the third alternative was the option of not choosing any of the previous ones. An example of a choice set is shown in Figure 1. Thus, each respondent was asked to choose the option according to his or her preferences and tastes for extra virgin olive oil.

2.2.2. Utility Model Specification

According to random utility theory [28], the utility U_ij for an individual i proportioned by a product alternative j can be decomposed into a deterministic (V_ij) and a stochastic (ε_ij) part, additively considered:

$$\begin{array}{c}{\mathrm{U}}_{\mathrm{i}\mathrm{j}}={\mathrm{V}}_{\mathrm{i}\mathrm{j}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{j}}={\sum }_{k_1=1}^{\mathrm{K}}{\mathsf{\beta }}_{\mathrm{i}{\mathrm{k}}_1}{\mathrm{X}}_{{\mathrm{k}}_1\mathrm{j}}+{\sum }_{k_1,k_2}{\mathsf{\beta }}_{\mathrm{i}{\mathrm{k}}_1{\mathrm{k}}_2}{\mathrm{X}}_{{\mathrm{k}}_1\mathrm{j}}{\mathrm{X}}_{{\mathrm{k}}_2\mathrm{j}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{j}},\\ \forall k_1,k_2=1,\dots ,K, k_1\ne k_2\end{array}$$

(1)

where V_ij represents the observed elements of utility determined by the k attribute levels (X_ikj), and ε_ij is a random error term with an independent and identically distributed distribution of extreme values [36]. Assuming a linear relationship between attribute levels, β_ik1 is the individual marginal utility obtained from each attribute k₁, which reflects how the level of utility changes with variation in the attributes of the extra virgin olive oil. However, the relationship among attributes is not always completely linear, and conjoint effects may arise among them. To capture these effects, interaction terms between pair of attributes (${\mathrm{X}}_{{\mathrm{k}}_1\mathrm{j}}{\mathrm{X}}_{{\mathrm{k}}_2\mathrm{j}}$) are also included in the definition of the utility function. As such, ${\mathsf{\beta }}_{\mathrm{i}{\mathrm{k}}_1{\mathrm{k}}_2}$ reflects the marginal effect on utility when attributes k_1, k₂ are both presented in product specification. The sign of the coefficient ${\mathsf{\beta }}_{\mathrm{i}{\mathrm{k}}_1{\mathrm{k}}_2}$ also shows the type of relationship between attributes: negative sign reveals substitution effects, whilst positive indicates complementarity.

The multinomial logit (MNL) model has been widely used to estimate choice probabilities [37]. However, this model operates under the assumption that individuals share the same coefficients, implying uniform preferences across population. To address this issue, the mixed logit (MXL) model allows marginal utilities to differ among individuals following a continuous distribution [36]. As such, preference heterogeneity is captured. Both MNL and MXL models are first estimated by assuming a linear specification for the utility function. Also, a second MXL model is estimated by including conjoint effects between the attributes Protected Designation of Origin (PDO) and ‘Territorio Sierra Espuña’ (TSE) brand. The MXL model is obtained by using the maximum simulated likelihood estimator with 500 Halton draws, and assuming a normal distribution for random coefficients.

From these models, respondents’ willingness to pay (WTP) for each attribute can be estimated as the marginal rate of substitution between the attribute k and the monetary attribute. It is determined by the following expression:

$${WTP}_{k_1}=-{\displaystyle \frac{{\beta }_{k_1}}{{\beta }_c}}$$

(2)

where β_k1 is the coefficient of any k₁ attribute, and β_c is the coefficient of the monetary attribute.

The results of the multinomial logit model were also used to calculate the relative importance of each attribute. From the estimated coefficients, it is possible to calculate the relative importance (Rimp_k) of the different attributes based on the rank-based and standardized weighting of each attribute [38]:

$$Rim{p}_k={\displaystyle \frac{{rank}_k}{{\sum }_{k=1}^K{rank}_k}}\cdot 100={\displaystyle \frac{{{\beta }_{max}}_k{- {\beta }_{min}}_k}{{\sum }_{k=1}^K{{(\beta }_{max}}_k{- {\beta }_{min}}_k)}}\cdot 100$$

(3)

2.3. Survey and Data Collection

Once the attributes and levels had been selected, and the choice experiment designed, a questionnaire was designed consisting of 20 questions divided into three sections.

The first section comprised questions related to the respondents’ pattern of consumption of extra virgin olive oil: frequency of consumption and where they usually buy this product. The second section presented the choice experiment. This is the central part of the questionnaire, from which to obtain social preferences for each of the attributes of extra virgin olive oil. The last section of the questionnaire included questions related to the sociodemographic and attitudinal characteristics of the respondents. Environmental attitudes were measured by means of affective, verbal and real environmental commitment indices, using a five-point Likert scale to evaluate a set of statements in the questionnaire [39].

At the beginning of March 2023, a pilot survey was carried out on 15 individuals, which allowed for refining the questionnaire. Ultimately, the final survey was conducted in person in April 2023 to a random sample among the 546,000 households in the Region of Murcia. A final sample of 416 surveys was obtained, which, for a binary variable and a confidence level of 95%, yields a sampling error of 4.8% for intermediate proportions and 2.9% for extreme proportions. It ensures the representativeness of the sample.

2.4. Sample Characteristics

Table 2. Sample description (n = 416).

Description	Mean	SD	Min	Max
Household size (people)	3.40	1.17	1	10
Age (years)	35.15	15.30	18	86
Family income (EUR/household/month)	2011	1192	0	6000
Personal income (EUR/person/month)	670	459.57	0	2500
Active worker (% yes)	52.16
University studies (%)	54.57
Gender (% women)	48.56
Affective ecological index (1–5)	4.18	0.86	1	5
Verbal ecological index (1–5)	3.52	0.98	1	5
Real ecological index (1–5)	2.15	1.14	1	5

shows the description of the main sample characteristics, identifying the profile of the average respondent as a man, 35 years old, with university studies (completed or in progress) and who is an active worker. The most prevalent household configuration is a four-member household (3.4 on average), with an average income of 2011 EUR/month. These characteristics do not differ significantly from the census values for the Region of Murcia [40], which confirms the representativeness of the sample and its validity for population analysis and inference.

The population displays a notable ecological commitment in the affective facet (4.18), intermediate commitment in the verbal facet (3.52), and comparatively low commitment in terms of real ecological performance (2.15).

3. Results

First, the pattern of consumption of ecological agri-foods and extra virgin olive oil is assessed.

Table 3. Evaluation of consumer behavior of ecological products ¹ and extra virgin olive oil.

Description	Mean
Households with high frequent consumption of ecological products (%)	18.03
Households with high frequent consumption of extra virgin olive oil (%)	78.84
Place of purchase of extra virgin olive oil (%)
Supermarket	66.11
Traditional shop	12.74
Directly from producer/self-supply	32.45
Online	1.92
Specialized shop (gourmet, eco, …)	5.05

¹ Ecological products include integrated pest management and organic agri-foods.

shows respondents’ consumption behavior, displaying a low frequency of purchase of ecological products. Conversely, most respondents exhibited a high consumption of extra virgin olive oil, predominantly from supermarkets or through self-supply.

Secondly, consumer preferences for purchasing one liter of extra virgin olive oil are assessed.

Table 4. Choice modelling results.

	Model 1			Model 2			Model 3
	MNL Linear Effects			MXL Linear Effects			MXL Conjoint Effects
Attributes	Coef.	SE	p-Value	Coef.	SE	p-Value	Coef.	SE	p-Value
Mean
Integrated	0.340	0.079	0.001	0.372	0.080	0.000	0.312	0.082	0.000
Organic	1.155	0.099	0.001	1.291	0.107	0.000	1.305	0.110	0.000
Protected Designation of Origin (PDO)	0.542	0.065	0.001	0.603	0.071	0.000	0.894	0.115	0.000
‘Territorio Sierra Espuña’ (TSE) brand	0.737	0.068	0.001	0.846	0.073	0.000	1.155	0.121	0.000
No purchase	−0.835	0.141	0.001	−0.992	0.151	0.000	−0.806	0.164	0.000
Price	−0.159	0.010	0.001	−0.172	0.011	0.000	−0.171	0.011	0.000
Conjoint effect PDO and TSE							−0.507	0.147	0.000
SD
Organic				0.591	0.128	0.000	0.613	0.129	0.000
Protected Designation of Origin (PDO)				−0.483	0.098	0.000	−0.487	0.097	0.000
No purchase				−0.172	0.011	0.000	0.636	0.112	0.000
n	416			416			416
Maximum likelihood	−2402.58			−2322.46			−2316.39
Adjusted R2	0.080			0.088			0.090
AIC	4817.15			4662.92			4652.78
BIC	4852.09			4715.19			4710.86

shows the results of the estimated models from choices of different alternatives of extra virgin olive oil according to their production system and geographical designations of origin. Model 1 presents an MNL-Linear specification, whilst Model 2 and Model 3 show an MXL-Linear and an MXL with conjoint effects, respectively.

Fixed coefficients of the MNL model indicate the marginal utility of each attribute and level. As such, organic extra virgin olive oil is the most demanded for the interviewed consumers, followed by that recognized by the ‘Territorio Sierra Espuña’ label. Protected Designation of Origin and integrated pest management production are also valued for the consumers, but with less intensity. The negative sign of the monetary attribute indicates that higher price levels provide greater disutility levels. Also, the ‘no purchase’ option reveals a negative for its coefficients, meaning that purchasing extra virgin olive oil is always demanded, regardless of its production system or geographical identification. Despite the meaningful results, the MNL specification considers homogeneous preferences across individuals.

Model 2 presents an MXL-Linear specification for extra virgin olive oil preferences. Only the attributes of organic production and protected designation of origin, together with the ‘no purchase’ alternative, show heterogeneous preferences. Their coefficients follow a normal distribution, only selecting those random attributes whose coefficients are statistically significant after a stepwise procedure. Model 2 performs better than Model 1, as confirmed by LR test (LR = 160.24; p-value = 0.00), AIC and BIC, revealing the existence of heterogeneity across individual preferences. The sign and effect of the mean coefficients for attributes of extra virgin olive oil are in line with those obtained for the MNL model. However, at this stage, it becomes important to highlight that preferences for the ‘Territorio Sierra Espuña’ brand are homogeneous across individuals and stronger than those disentangled by the MNL model—as its coefficient shows a greater mean value. As such, these results confirm the demand for locally produced extra virgin olive oil.

The relationship among attributes is not always linear and additive. Model 3 extends the results of Model 2 to encompass the conjoint effects that may arise between labels of product origin when they both are presented in extra virgin olive oil: Protected Designation of Origin and ‘Territorio Sierra Espuña’ brand. The coefficient of the interaction term between these two attributes reveals a negative sign, meaning that their relationship is substitutive. That is, when both labels of certification of origin are presented in the same product, their conjoint effect is less than the sum of their isolated effect. This also means that the presence of a certification label of origin would compensate, to some degree, for another certification label. Additionally, the results of Model 3 also reveal that the isolated effect of both Protected Designation of Origin and ‘Territorio Sierra Espuña’ brand are greater than those revealed in previous models, showing stronger preferences for the presence of, at least, a certification label of origin. Model 3 performs better than Model 2 according to main selection model criteria—LR test, AIC, BIC—and so it will be used for further assessment.

Respondents’ willingness to pay (WTP) for each attribute is also estimated and shown in

Table 5. WTP for extra virgin olive oil attributes.

Attributes	WTP (EUR/L)	Confidence Interval (95%) 1
Integrated	1.82	0.91	2.73
Organic	7.35	7.01	7.68
Protected Designation of Origin (PDO)	5.18	4.93	5.43
‘Territorio Sierra Espuña’ (TSE) brand	6.74	4.88	8.60
Conjoint effect PDO and TSE	−2.95	−4.71	−1.20

¹ Note: Confidence interval for non-random attributes is estimated by using the Delta method.

. According to the WTP for each attribute, respondents are willing to pay a surcharge of EUR 7.35 for a liter of organic extra virgin olive oil; EUR 6.74 for a liter of ‘Territorio Sierra Espuña’ brand extra virgin olive oil; EUR 5.18 for a liter of extra virgin olive oil recognizing Protected Designation of Origin; and, finally, respondents would pay EUR 1.82 extra for a liter of extra virgin olive oil obtained through an integrated pest management system. Additionally, the substitution effect between Protected Designation of Origin and ‘Territorio Sierra Espuña’ brand is valued at EUR −2.95 for a liter. As such, the WTP for a liter of extra virgin olive oil identified with the labels of Protected Designation of Origin and ‘Territorio Sierra Espuña’ brand arises to EUR 8.97.

In addition, respondents were asked to indicate the reasons why they would buy extra virgin olive oil from the ‘Territorio Sierra Espuña’ brand. The most repeated reason was ‘to contribute to the local economy and rural development’, chosen by 68.27% of respondents, followed by ‘because it is a healthier product, with less pesticides...’ (57.45%) and ‘to contribute to the protection of the natural environment’ (46.63%).

The choice experiment results also allow the relative importance of each of the attributes of extra virgin olive oil to be calculated (Figure 2). Respondents’ choices reveal that the production system (47.12%) is the most important attribute to consider when purchasing extra virgin olive oil. Of the two quality certification attributes, the ‘Territorio Sierra Espuña’ brand (30.86%) is much more important than the Protected Designation of Origin (22.02%).

4. Discussion

The present study explored consumers’ purchasing preferences for agri-food products by combining the joint effect of the agricultural production system and origin certification labels. This has received little attention in the literature from a socioeconomic point of view. To address this gap in the literature, a choice experiment exercise was applied in this study to the case of the extra virgin olive oil product in a protected geographical area of south-eastern Spain (Region of Murcia). In this area, ecological oil production systems are being promoted in conjunction with local quality certificates (‘Territorio Sierra Espuña’ brand), seeking to join local and sustainable development.

The results indicated that both the agricultural production system and quality certification labels increased consumer demand for extra virgin olive oil, although with a different order of priority between them. It is therefore evident that the findings of this study demonstrated that consumers exhibit a higher propensity to purchase and a greater WTP for extra virgin olive oil when it is produced using an ecological system, either organic or integrated pest management system. It is important to consider that the word ecological has acquired a powerful meaning in today’s market. When consumers read this term in products such as olive oil, they not only associate it with environmentally friendly cultivation and production practices but also with a healthier and more natural perception of the product. As it stands, the results obtained are consistent with the extensive body of literature that demonstrates consumers’ greater preference for organic systems over more conventional production systems [41,42], and even over integrated pest management systems [43]. Also, they corroborate the market strategies generally followed by supermarkets, with surcharges ranging from 50 to 100% for organically produced extra virgin olive oil over conventionally produced extra virgin olive oil, which serves even as external validation for the choice experiment.

Regarding the influence that the selected certificate of origin labels have on consumers’ choices, the estimates indicated that all of them have a positive and statistically significant effect, albeit with differing magnitudes. This suggests that the presence of a certificate of origin label on extra virgin olive oil is associated with a higher utility compared to the alternative without any label, implying that when an olive oil is provided with quality certificates, its probability of being bought increases. Consumers associated greater utility with extra virgin olive oil of local origin through the ‘Territorio Sierra Espuña’ brand than with a more national projection, such as the Protected Designation of Origin label. These results are in line with the conclusions of other studies [44] on consumer preference for locally sourced oil in Italy compared to European origins, and in studies in the same study area [26] on the willingness to pay a higher price for ‘Territorio Sierra Espuña’ brand honey. Even then, market research for Protected Designation of Origin surcharges prices from 30 to 90% over conventional extra virgin olive oil, externally validating the results here obtained.

Heterogeneous preferences for organic and Protected Designation of Origin were revealed. Despite the great demand for extra virgin olive oil produced under organic systems at Protected Designation of Origin, it is recognized that consumers’ positive preferences for these attributes may significantly vary among individuals. This is in line with the heterogeneous preferences for organic attributes and local certifications revealed in different studies for olive oil [22] and other food products [20,45]. Such results clearly contrast with preferences for alternative ecological production systems, and, mainly, with the ‘Territorio Sierra Espuña’ brand, for which positive and homogeneous preferences were shown. Hence, following the purpose of this work, this clearly shows that local production systems under widely identified systems are demanded homogeneously across the population.

Non-linear preferences for certifications of origin were evidenced. The presence of different labels for certification of origin, as proposed here using the Protected Designation of Origin and the ‘Territorio Sierra Espuña’ brand, raises the WTP for extra virgin olive oil. However, the inclusion of more than one label exacerbates, to some degree, an overlapping effect between them, meaning that—although both are more valued than only an isolated certification—the increase in their value is not as high as expected from their respective isolated WTP. Some substitution effects therefore arise between origin certification labels, revealing the presence of diminishing marginal utility. These effects have been previously investigated in the literature, showing also the substitutive relationship among attributes regarding organic production systems and local certifications of origin [45], but not regarding different types of certifications of origin. It broadens the scope for future research seeking an in-depth understanding of the relationships among food attributes.

The greater relative importance of the agricultural production system with respect to quality labels of origin is in line with other research that concludes that quality labels are less important for consumers when ecological signals are present [10,46], and there is even a negative interaction between product origin and ecological labeling [47]. Furthermore, greater relative importance is observed for local rather than national geographical certificates, which also adds to the literature, revealing that nearer and concrete production places are preferred over national and international ones [20,48]. However, when the focus is on olive oil, the literature does not provide conclusive statements. While some works reveal that product origin dominates over organic production systems [20,32], others conclude that the contribution of olive oil production to biodiversity and ecosystems is valued greater than its contribution to local and rural development [22]. Hence, our research provides some light to the debate supporting the prevalence of production systems over origin recognitions.

The results suggest that the integration of information related to the production system and origin certification labels in marketing strategies could effectively increase the perceived value of agri-food products. Furthermore, this study supports the voluntary framework of the new EU Regulation 2024/1143 on ecological products and geographical certificates for producers to apply sustainable practices [33], which could contribute to market differentiation and sustainability of agroecosystems. While more information is needed to complement the work done, such as preference heterogeneity analyses, this research provides valuable insights for future research on how information related to the production system and geographical quality certificates affects consumer behavior in purchasing contexts.

Not only might more information about the labelling of agri-foods be required to capture the consumer surplus of local and organic products, but also their distribution needs to be considered. Alternative distribution channels, such as online platforms, social media, showrooms, and crowd shipping, among others, are being proposed to maximize consumer reach and service, also for the olive oil market [49]. Short food supply chains, which minimize intermediaries between producers and consumers, also emerge as sustainable alternatives, appealing especially to consumers who value local and organic products and are willing to pay a premium for them [34]. However, as the results reveal, supermarkets are the main channel for consumers to purchase extra virgin olive oil, which also adds to the consumer reluctance toward online purchasing [34]. Ensuring that local and organic products reach end consumers, therefore, means considering their availability within large-scale distribution channels.

Agri-food labelling should address also the supply perspective. Quality labels of geographic origin require farmers to adopt certifications that verify their compliance with management and production requirements. This is exacerbated when referring to the production systems, especially for the case of organic systems, whose requirements in terms of agricultural practices are greater. The adoption of these certifications by farmers might imply some transaction, investment and operational costs that make farmers reluctant to adopt them. Alternative incentives, such as subsidies and operational funds, may arise as policy options to ensure farmers actually adopt the requirements for certified local and organic production [50]. Hence, not only is it needed to ensure great demand and willingness to pay a premium for locally and sustainably produced agri-foods, but also to be able to provide them at competitive prices for suppliers.

In this sense, this study can lay the foundations for defining a conceptual framework for future research with which to carry out a real choice experiment in real stores that will allow a clearer reflection of consumer preferences [51]. Likewise, in order to make consumers feel more familiar with quality certificates from local geographical areas, and therefore, more known by the general population, it is recommended to market these type of products beyond the geographical area where these products are produced, which would help to have a greater social perception for this type of territorial brand that promotes both the conservation of the environment and the sustainable socioeconomic development of local economies.

One of the limitations of this study resides in its regional scope, directly derived from the type of certification labels of origin assessed. Despite this, it is quite common in the literature [22,26,39], and future research could study whether the conclusions of this study could be extended to other regions, or even replicate it with a nationwide perspective, so that assessment of the influence of distance decay could also be assessed. In addition, next steps would also be extended to other products with different types of local and ecological quality certificates, providing valuable information for the study of consumer preferences in different geographical or cultural contexts. Furthermore, other attributes significant for understanding consumer choices would need more attention in research (e.g., taste, package, environment certifications…), so that consumer behavior for extra virgin olive oil is comprehensively assessed. These approaches would improve the understanding of how information related to the type of production, their branding and organoleptic characteristics, and their interaction between different certificates applies in different markets and product categories, offering broader implications for producers and policy makers.

5. Conclusions

The present study makes a significant contribution to the existing literature by providing novel evidence derived from an agri-food choice experiment. This experiment integrates attributes pertaining to agricultural production systems and certificates of origin labels. In selecting extra virgin olive oil as a case study in a protected geographical area in south-eastern Spain, it was revealed that consumers’ choices to purchase this product were primarily influenced by the type of production system, with an evident preference for organic systems, followed by integrated pest management alternatives. Consumer preferences for local certificates of origin were also confirmed, which highlights the need to extend the sales of these products beyond the geographical limits where they are produced in order to achieve greater market success. This demand for organically and locally produced extra virgin olive oil is quantified with an average WTP of 7.4 EUR/L and 6.7 EUR/L, respectively, which represents the maximum surcharge stated by consumers regarding conventionally produced extra virgin olive oil without any origin certificate. Also, substitution effects between certificates of local origin arise, revealing the importance of cautiously selecting those certificates deeply related to production sites.

Therefore, including information on the type of production system and local labelling on agri-food products can be a useful marketing strategy for producers who are inclined towards environmentally friendly production, while at the same time satisfying social demand for quality agricultural products.

These results suggest an in-depth theoretical analysis about complementary and substitution effects among production systems, local certifications of origin and any other significant attributes for consumer choices of extra virgin olive oil. Furthermore, mechanisms to ensure suppliers are willing to implement the required production systems and certifications may also become a challenge for future research avenues.