Sustainability, Corporate Social Responsibility, and Performance in the Spanish Wine Sector

: This paper analyzes the Corporate Social Responsibility (CSR) of a sample of wineries in Spain and its effect on the companies’ performance. We used a questionnaire created with a validated scale that includes environmental and social dimensions. The ﬁnal sample was made up of 127 ﬁrms that participated in the National Wine Fair (FENAVIN) in 2019, and this was analyzed using a cluster analysis and means contrast to verify whether a CSR-performance relationship exists. The performance was measured using the average return on assets (ROA) for the prior three years. The results obtained do not support the general opinion that enterprises involved with CSR achieve better results. On the contrary, we found that wineries that are more environmentally responsible are the least proﬁtable and that those with more socially responsible behavior do not have a signiﬁcant CSR-performance relationship.


Introduction
The origin of the concept of sustainability lies in the well-known definition of sustainable development established by the United Nations in 1987. In summary, sustainability permits people to live as they want or need without spoiling the future of their fellow human beings. Although it is sometimes related only to the environment, sustainability is closely linked to Corporate Social Responsibility (CSR), as shown in specialized literature [1,2], since CSR allows companies to express their commitment to sustainability. It is also broadly recognized that the concept has a triple dimension: environmental, social, and economic.
Many studies have analyzed the influence that the transition toward more sustainable production processes has on production costs and profitability, or how critical it is for companies to conciliate environmental and social goals with economic objectives. Research such as [3] claims that the integration of CSR practices into firms allows them to improve the value of the brand, access to financing, motivated employees, public image, and loyal customers, and to, in general, enhance the relationship with stakeholders. However, we consider that a research gap exists since the results obtained from different studies have not been conclusive. The principal objective of this research is, therefore, to verify the potential relationship between profitability and sustainable policies within the Spanish wine sector distinguishing, at the same time, between the environmental and social dimensions of CSR. The literature review leads us to affirm that most of the studies confirm the existence of a positive relationship between CSR and profitability [4][5][6][7] and therefore that is the hypothesis that we also propose in this research.
We chose the wine sector because of its importance for the economy in Spain. A gap between production and domestic consumption has been observed in recent years, which has led Spain to become the country with the largest volume of wine exports, with a global market share of 20.5% [8]. Spanish wineries have transformed their production processes and included the latest technologies and improvements, investing more than 800 million euros in this modernization of the sector [9]. Some researchers have proved that when specific resources are assigned to sustainable activities in the winery, the adoption of such processes does not damage the company's profitability or substantially increase their costs [10].
In order to verify the potential CSR-profitability relationship in the wine sector, we developed a cluster analysis using a sample of Spanish wineries. The main results obtained from this research can be summarized as follows: we found that more environmentally responsible wineries are the least profitable, while an analysis of socially responsible wineries showed that there was no significant CSR-performance relationship.
This paper is structured as follows. We first describe the literature review carried out to discover research concerning sustainability and CSR concepts, their relationship with business performance, and their situation within the wine sector. We then go on to explain the methodology used to attain the data and the empirical analysis. Finally, we present the main results and the conclusions that can be drawn from them.

Sustainability and CSR
Sustainability and CSR are currently major concerns in the corporate world. There has been a change of social conscience in the sense that firms should not obtain profit at any costs and should keep in mind the potential impact that their activities have on the environment. Both terms are connected: "CSR is the responsibility that companies have for their impact on society (social, economic, and environmental) and, therefore, seeks to minimize negative impacts and maximize positive ones. Sustainability refers to the company's ability to meet its needs without compromising future generations" [11]. Some research collecting common definitions of these two concepts has concluded that they are very closely connected, although CSR is usually defined more normatively. The main objective of CSR is to improve companies' behavior, leading them toward sustainability [12]. According to [13], which provides a very integrative definition of CSR, a socially responsible company is one that develops a profitable activity that takes into account all the positive and negative economic, environmental, and social effects that it may have on society. CSR is considered an indicator of the success of a firm as a whole and a potential means to obtain sustainable development. Adopting CSR as a fundamental strategy allows enterprises to not only satisfy their stakeholders but also contribute to the firms' sustainability [14].

CSR and Performance
Several studies have concluded that companies that develop CSR actions obtain a better performance [15][16][17][18][19][20][21]. Companies involved with CSR issues might obtain recognition that can take the form of better access to financial resources, higher consumer demand, and an increase in worker productivity. They might also improve their reputation and attain the appreciation of the social community, thus enhancing their value in the long term [18]. However, other research works have found no significant relationship between CSR and performance, or have even concluded that CSR actions damage the firms' performance [22,23]. Some of them have gone so far as to claim that being socially responsible is not appropriate for companies, since it is public institutions that should offer this type of services. Friedman [24] claims that enterprises should only respect the law without forgetting that their main objective should be to maximize the companies' profits. That is, CSR activities represent a voluntary involvement on the part of the firms that goes further than the limits of legislation and the main goals of enterprises. CSR activities also cost money, which can lead to a decrease in performance among responsible firms [25].
According to [26], which comprises a review of the main meta-analyses regarding this issue, a positive relationship between CSR and profitability was found in the majority of the research papers studied. For example, [27] found 54 papers reporting a positive relationship, 28 reporting a non-significant relationship, and 7 reporting a negative relationship. The authors of [28] claim that 65 percent of the papers reviewed showed a positive result, 31 percent a neutral or non-existent relationship, and 19 percent a negative relationship. Accordingly, the positive link between CSR and profitability has been validated by many empirical studies. Therefore, the two hypotheses below are proposed, distinguishing between the environmental and social dimensions of CSR.
The adoption of CSR by companies can have three effects: output, outcome, and impact [29,30]. Output is the engagement and strategies with regard to introducing CSR concerns into business processes and their relations with stakeholders (for example, publishing a code of ethics or being included in the sustainable stock market indexes). This output generates some types of behavioral improvement within firms that is denominated as the outcome. All of this must produce effects for the environment and society, which is called impact. An enterprise's CSR output is usually publicly available and, therefore, easy to obtain. However, in order to show a company's outcome, the use of questionnaire data or interview information is required. The company's CSR impact can eventually be measured using certain items such as the reduction in energy consumption or the use of plastic containers. In this paper, we have focused on the enterprises' outcomes and have, therefore, employed a questionnaire to obtain information concerning firms' social and environmental actions.

Sustainability and the Wine Sector
Spain is the largest wine producer in the world. It produces far more wine than other traditionally important producers such as South Africa, Chile, or Argentina. Both Spain and California, in the USA, are the areas with the main surface area and production of wine worldwide. A study of the wine-producing regions in Spain reveals that they have several differences in relation to this issue. Rioja and Ribera del Duero, the northern area of the country, have higher added value and margins when compared to the raw material, with a higher gross operating margin. Castilla-La Mancha is, meanwhile, the most productive area, but obtains narrower gross operating margins than the regions in the north. It focuses more on bulk wine and lower prices, which has permitted it to achieve export dynamism. At the end of nineties, wineries benefited from high prices in world markets, but from the 2000s to the beginning of the 2008 worldwide crisis, the wineries' high investments and excessive personnel costs reduced the firms' profits. New companies that had not even a minimum level of knowledge about the sector were also created. However, from 2008 to 2011, at the height of the economic crisis, Spain was able to achieve a great growth rate in volume and conquered international markets [31].
If we focus on the food industry, it is difficult to find a unanimously accepted definition of sustainability, although a general idea has extended that covers a wide variety of aspects, such as minimum exposure to chemicals, animal welfare, the environment, and workers' rights [32]. It is, moreover, not easy to define this concept in the wine sector because each country and each winery has a different interpretation of the term. It has traditionally been linked principally with the environmental issues of wine production, whilst other fundamental aspects such as workers' rights or community development have been ignored [10].
The triple dimension of sustainability first appeared in the wine sector in 1999, as can be seen in the last resolution of the International Organization of Vine and Wine (OIV): the 2016 General Principles of Sustainable Vitiviniculture, in which environmental, economic, and social aspects are combined [33]. The wine sector is affected by sustainability issues in several respects, and there are several problems, such as exposure to chemicals, climate change, and the lack of water and energy. A sustainable strategy could, therefore, be a competitive resource, a driving market factor, and a fundamental driver for the innovation process [34].

Methodology
The sample was obtained from the Spanish wineries that participated in the National Wine Fair (FENAVIN) in 2019. After nine editions, this has become the most important monographic trade fair in Spain. It takes place every two years and constitutes an important reference for Spanish wineries, with the presence of firms representing most of the country's Designations of Origin.
A total of 1946 firms attended FENAVIN, and their managers and CEOs filled in the questionnaires during the three days that the event lasted. After excluding companies with missing data from the sampling, we eventually obtained 127 valid questionnaires (Table 1). The questionnaire was created after an extensive review of the relevant literature and using a scale validated in [35]. We focused the analysis on the environmental and social dimensions of the CSR, which were evaluated using ten items ( Table 2). Each item was measured using a 5-point Likert scale, on which the values were classified from 1 (completely disagree) to 5 (completely agree), signifying that the higher the score, the higher the wineries' implications with environmental and social issues. Table 2. Scale of the CSR orientation.

Envir1
We are able to minimize our environmental impact Envir2 We consider energy savings to achieve higher levels of efficiency Envir3 We consider the introduction of alternative energy sources worthwhile Envir4 We have a positive predisposition to use, purchase, or produce ecological items Envir5 We assess the use of recyclable packaging

Soc1
Employees' wages are linked to their skills Soc2 We are committed to job creation Soc3 We encourage the training and professional development of employees Soc4 We have policies regarding labor flexibility Soc5 We consider employees' proposals in management decisions Source: adapted from [35].
We applied two cluster analyses to the data attained, one to the environmental dimension and another to the social dimension. The main purpose of this statistical test was to classify a sample into groups that are as homogeneous as possible in relation to the observed variables. We specifically wished to group the companies in our sample according to their different degrees of commitment to CSR. We subsequently applied the comparison of means in order to verify whether the differences among the groups' return on assets (ROA) were significant.
There are several types of performance measures, such as profitability, asset utilization, growth, and market-based ratios [36]. The first includes return on investment, return on equity, return on sales, profit margin, and earnings per share. The second includes asset turnover and return on assets, while the third includes average profitability and average asset utilization. The last includes measures such as market value to book value or market return. We chose the growth measures because they would allow us to include information concerning several periods of time. In this research, therefore, the mean ROA for the years 2016, 2017, and 2018 was used to develop the test. This measure was also chosen in other research related to the wine industry: in the French wine-producing sector [37]; in wineries from California, Italy, and India [38]; and in the Spanish wine sector [39]. We obtained the ROA of each company from the Iberian Balance Sheet Analysis System (SABI) database.
When applying the cluster analysis, it is necessary to decide two things: the measure of the distance between individuals and, once this has been defined, the criterion employed for the formation of the clusters. Regarding the first question, since the number of individuals and variables was small, the Euclidean distance was the most appropriate measure. Regarding the criterion for the formation of clusters, we used the agglomerative hierarchical cluster analysis and employed the average linkage between groups to link the clusters. This made it possible to follow the process stage by stage, obtaining the number of groups a posteriori.

Results
An analysis of the dendrogram obtained allowed us to conclude that the most appropriate option was that of two clusters, since it was the solution in which the clusters were distant from each other while the elements within each one were close together. We, therefore, considered these two groupings and the means obtained for the ROAs in order to analyze the differences in means.
The main results obtained when considering the environmental variables (Hypothesis H1) were the following: Cluster 1 attained higher scores than Cluster 2 in the five variables (Table 3a). There were significant differences between the mean ROAs in the two clusters with a confidence level of 90%, and the ROA mean ratio of Cluster 2 is superior to that of Cluster 1 (Tables 4a and 5a).   The main results obtained when considering the social variables (Hypothesis H2) were: Cluster 1 attained higher scores than Cluster 2 in the five variables (Table 3b). There were no significant differences between the mean ROAs in the two clusters. In this case, the ROA mean ratio of Cluster 1 is superior to that of Cluster 2 (Tables 4b and 5b).

Discussion and Conclusions
The economic impact for those enterprises undertaking CSR projects remains unclear. The conclusions obtained by the various research works studying the impact of CSR on firms' performance are indefinite. While some claim that CSR activities and processes could increase an enterprise's profit [26,40,41], others deny this [42,43] or discover that the influence is either difficult to measure or neutral [44,45]. The results obtained in this paper do not support the general opinion that enterprises involved with CSR achieve better results and, therefore, neither do they support the Hypotheses H1 and H2 that we raised. On the contrary, we found that more environmentally responsible wineries are the least profitable and that those with more socially responsible behavior do not have a significant CSR-performance relationship. The first result is in line with the so-called trade-off hypothesis first proposed by [24]. The superior costs produced by CSR could be detrimental to the company's performance and the managers' main objective is to increase the firm's profit. The second result is in line with the so-called moderating variables hypothesis [46], i.e., it is difficult to discover a relevant and significant link between CSR and performance. Other variables, such as the firm's activity, size, age, investment in innovation, etc., may participate in this relationship [47]. In this respect, [48][49][50] developed research showing a broader relationship between CSR and competitive success, including other items beyond those of an accounting, financial, or economic nature.
As an additional conclusion, we should also state that the behavior shown by the wineries analyzed in relation to CSR is quite uniform. There is a notable difference in the number of companies belonging to each cluster. The fact that most companies belong to the first cluster indicates that the responses to the questionnaire were similar. This leads us to deduce that there is a similar strategy among wineries in relation to this issue.
Analyzing the managers' responses, we can conclude that most wineries are involved in environmental issues, particularly as regards employing practices related to the responsible use of energy and recyclable products. In the social dimension, the majority are equally involved with the well-being of employees through fair wages, professional development, flexibility, and participation, as reflected in the high scores obtained for those items in the questionnaire.

Limitations and Future Research Suggestions
This paper, therefore, makes several contributions that may be of use to managers and scholars, but also has some limitations. It is necessary to keep in mind that the mean ROA, which was chosen to measure business performance in this research, is usually used to represent short-term economic results. Intangibles are an important component of the CSR and have long-term effects on business performance. The majority of authors claim that CSR actions favor the growth and sustainability of enterprises [51]. In order to complement this research it might, therefore, be advisable to use market-based measures that are able to capture the long-term influence of CSR [52]. The results of this paper are similar to those obtained by [46,53], whose research showed that CSR does not influence business performance in the short term.
The means employed to measure CSR can also be criticized, since there are many possibilities, some of which are more objective: pollution ratings, corporation philanthropy, annual report disclosure, position in different ratings related to CSR, compliance and responsive behavior, etc. Moreover, the potential influence of some other variables, such as the sector, size, or age, may moderate the causal relationship between CSR and ROA. All these options may be objects of future studies that could contribute to clarifying this issue. In the questionnaire we used to carry out this research, we added some questions related to the company's competitive strategy. In a future study we will analyze the mediating effect of this variable, which may also have important effects.