IntroductionTop

Case study

The Cava PDO is the largest bottled wine exporter in Spain, slightly ahead of the Rioja PDO, and accounts for around 27% of the total national wine export volume (MAPA, 2020). Fig. 1 shows the evolution of both PDOs in the last 20 years.

The history of Cava begins with the phylloxera epidemic that devastated French vineyards in 1870, which represented an opportunity for the entire Spanish wine industry to become a supplier to the neighboring country (Fernández & Pinilla, 2014). It was also the beginning of Josep Raventos’ research with the traditional champagne method, with indigenous grapes from Catalonia, particularly parellada, macabeo and xarel-lo varieties. Since 1890, Cordorniu, Josep Raventos’ winery, decided to specialize in the production of the sparkling wine that would later receive the name of cava (Saito & Takenaka, 2004). When phylloxera destroyed the Penedès vineyards at the beginning of the 20th century, most of the farmers took the opportunity to substitute the traditional red grapes for the varieties that Josep Raventos experimented with to make cava (Saito & Takenaka, 2004). The subsequent protectionist policies, which made it difficult for French champagne to enter the Spanish market, and the advertising techniques of Manuel Raventos, son of Josep Raventos, facilitated the establishment of a large production center in the Penedès region. After this initial settlement came a subsequent export stage, in which the Freixenet brand played an important role, positioning cava as the most important wine in terms of export volume (Fernández & Pinilla, 2014). In 1972, the Regulatory Council for Sparkling Wines was established, and the name “Cava” was approved to denote Spanish sparkling wine (CRC, 2021a).

One unique characteristic of the Cava PDO that other PDOs do not share is that wineries belonging to the Cava PDO that make sparkling wine are able to be located out of the original Penedès production area, although 90% of the Cava PDO wineries belong to Penedès (Saito & Takenaka, 2004).

The area dedicated to Cava was 32,009 ha in the 200001 campaign, compared to 57,636 ha in Rioja and 673,626 ha in all PDO regions of Spain. The data for the 2018-19 campaign reflect an area of 37,955 ha for Cava, 66,239 ha in Rioja, and 648,631 ha in the national total. Across this 19-year period, we can therefore observe a growth of Cava of 18%, a growth of Rioja by 15%, and an overall decrease in total PDOs of 3.7% (MAPA, 2020). The data of the number of winegrowers, wineries, volume of qualified wine, volume of wine sold in domestic market, and abroad, can be seen in Table 1. Table 1 shows that there has been a different evolution in the Cava PDO, with two variables that go against the general evolution of the sector: an increase in the number of wineries and a decrease in the volume sold in the domestic market.

In the 19-year period from 2000 to 2019, the number of wineries increased by 84 in the Cava PDO, the agricultural area by 5,946 ha, and the volume of qualified wine by 332,478 (MAPA, 2020). Cava’s increase in the number of wineries, contrary to Rioja and the PDOs of Spain as a whole, has caused a lower average volume of qualified wine per winery, which has gone from 6,673 hL in 200001 to 6,034 hL in 2018-19. This output is still well above the ratio of Rioja, which is around 4,442 hL per winery in 2018-19, as well as the national average of 2,891 hL per winery (MAPA, 2020). This process, inverse to that which has occurred in the Spanish wine sector, can perhaps be explained, on the one hand, by the interest that Cava has aroused in regions that did not produce it, and by a lesser need for investment for its production, which has allowed small producers to quickly enter the market, linked to the Masía concept as a differentiated production core (Saito & Takenaka, 2004).

The prices of Cava wines in this period have had an interesting evolution, rising in the domestic market, and decreasing in the foreign market. Table 2 shows the data for this period from 2000-01 to 2018-19.

The variation of prices and quantities in the domestic and foreign markets can be seen in more detail in Figure 2. The trend line for the data has been added to them. The trend line allows us to calculate the elasticity of domestic demand. For its calculation, an intermediate point of the domestic market P = 500 €/hL and Q = 700,000 hL has been taken, resulting in an elasticity of -0.60. The trend line shown in Fig. 2 also allows us to calculate the elasticity of external demand; for this we have taken an intermediate point P = 300 €/hL and Q = 1,000,000 hL, resulting in an elasticity of -30.

Therefore, domestic demand, in absolute value, is inelastic (< 1) and determines that the income of producers decreases as the quantity exchanged increases. However, foreign demand is elastic (> 1) in absolute value and suggests that the income of the wineries increases as the exported volume increases (Krugman et al., 2015).

An element that links the wine activity with the production area and its sustainability is the remuneration of the resources used in production (Simon-Elorz et al., 2015; Villanueva & Ferrer, 2020). Table 3 makes a comparison of Cava PDO, Rioja PDO and the mean of PDOs in Spain; due to a lack of complete data, the period from the 2016-17 season to the 2018-19 campaign is analyzed. The data show that the values of the resource per hectare and per winery present higher data in Cava than in Rioja and in the average of the PDOs of Spain.

 

Figure 1.  Total wine exports by volume from PDO Cava and PDO Rioja, 2000-01 to 2018-19. Source: Author’s own elaboration based on MAPA data (2020), series 2000-01 to 2008-19.

 

Figure 2.  Price-quantity graph. Domestic and Foreign Cava market, from 2000-01 to 2018-19. Source: Author’s own elaboration based on MAPA data (2020), series 2000-01 to 2008-19.

 

Table 2.  Changes of the value, quantity, and price of Cava PDO wines in the domestic and foreign market.

Source: Author’s own elaboration based on MAPA (2020) data, series 2000-01 to 2008-19.

 

 

 

Resources and capabilities or resource-based view

This first theory focuses on the availability of certain resources and capabilities in the company, the key element to achieving a competitive advantage (Barney, 1991). Re-sources are all available factors that the business controls and that are turned into end products or services using a wide range of other assets and mechanisms available to the business. Capabilities develop over time, based on complex interactions between available resources (Amit & Schoemaker, 1993). The resources and capabilities available to the company are not in themselves a strategic and fundamental element that ensures the achievement of a competitive advantage. To obtain it, three conditions must be met: a company must establish the competiti-ve advantage, maintain the competitive advantage, and appropriate the benefits derived from it. To satisfy these three conditions, resources must be scarce and relevant, and they must be durable, non-transferable, and not repli-cable (Grant, 2010).

 

Strategy

For the company, a strategy is a plan that determines the objectives (Brenes et al., 2014), integrates policies and the most relevant sequences of action (Mintzberg et al., 2009), and contemplates which markets to supply and with which products (Ansoff, 1965). Despite being criti-cized for the overly static nature of his approach or his failure to account for evolving components of company’s real strategies (Mintberg et al., 2009), Porter’s model con-tinues to be the reference model for analyzing business strategy (Brenes et al., 2014; Islami et al., 2020). Porter (1985) argues that to obtain a competitive advantage, a company must choose between two strategic options, lea-dership in cost or leadership in differentiation. Cost lea-dership focuses on the production of low-cost products to satisfy price-sensitive customers, while leadership in differentiation focuses on offering different and unique products and services in the industry, but to a wide range of clients that are relatively insensitive to price (Soltani-zadeh et al., 2016). Both strategies can be successful, and the literature indicates successful situations for both. The differentiation strategy has been referred to as the star stra-tegy; in the context of the food industry in Italy, Curzi & Olper (2010) argue that differentiation strategies influence the ability of companies to develop their external activity. In addition, based on business activity in Colombia, Kugler & Verhoogen (2008) reveal that to carry out good quality practices, highly qualified employees are needed, establishing a positive relationship between investment in inputs and the quality and price of products. However, reality shows many examples of companies that increase their presence in the international market through cost strategies (Bardají et al., 2014; Simon-Elorz et al., 2015). In the Spanish wine sector, faced with the challenge of internationalization due to the decline in domestic consumption, wineries have been able to combine increases in value and new customers with efficient cost-reduction models that have resulted in very competitive prices (Cervera Ferrer & Compés López, 2018).

Although Porter’s model has a preponderant role in the definition of strategies through the two classics of differentiation and costs, there are different models that broaden this vision and introduce more types of generic strategies. Among them is the Robinson & Pearce (1988) model, which defines four generic strategies—efficiency, service, innovation, and marketing—and which has been used in different studies. For example, the Robinson & Pearce model has previously been applied by Spanos & Lioukas (2001) in Greek manufacturing industries, by Salavou & Sergaki (2013) in Greek private food firms, by Brenes et al. (2014) in Latin American agribusiness, by Ferrer et al. (2018) in the Spanish wine sector, and by Villanueva & Ferrer (2020) in the U.S. wine sector. It is necessary to mention that this expansion of Porter's classic model, towards a model that includes in the company's strategy a greater compatibility of generic strategies, where innovation, efficiency, and marketing are combined, for example, is at the base of the Blue Ocean strategy (Kim & Mauborgne, 2005). That same argument is defended by D’Aveni et al. (2010), in mature markets characterized by strong competition such as that characterizes the wine sector.

 

Table 3. Production value in Cava PDO, Rioja PDO, and mean of all Spanish PDOs.

Source: Author’s own elaboration based on MAPA data (2020), series 2016-17 to 2008-19.

 

 

 

 

Managerial capabilities

Management and organizational capabilities are developed at the top of the organizational chart through three functions: coordination and integration, learning, and reconfiguration (Teece et al., 1997). These capabilities are part of the routines learned and differentiate the company, explaining why some of them present a more efficient management than others and become a source of competitive advantage (Teece et al., 1997).

The importance of managerial capabilities is based on the manager’s vision and leadership (Pickett, 1998), integrating this with the strategy (Westley & Mintzberg, 1989). The managerial competencies include the definition of the strategy and the organizational structure at the level of design and implementation. Managers must provide a high degree of commitment, clear definition of objectives and financial resources (Pickett, 1998), and guide employees towards the shaping of business resources and competencies (Kor & Mesko, 2013).

Management resources reflect the capabilities of managers and are precursors to competitive advantage and performance. The managerial capabilities are not easily exportable to other companies. Therefore, the hiring of external managers does not always have a positive effect. It is through human capital that the manager generates income by implementing strategy and making operational decisions (Castanias & Helfat, 2001; Helfat & Martin, 2015).

Managers use their management capability to guide the company towards cost reduction, product differentiation, or a combination of both, looking for competitive advantage. Their management responsibilities include strategic business vision, internal communication, strategic management of human resources (recruitment, job analysis, development, training, performance, and compensation), the acquisition of resources, and their transformation into products and services. Through these managerial steps, they create value for the partners and owners of the company thus being a generator of revenues and their appropria-tion, and a key element for the maintenance of the compe-titive advantage (Lado & Wilson, 1994). The analysis of managerial capabilities and their connection with strategy and performance has been analyzed, finding a direct re-lationship between the former and the latter (Spanos & Lioukas, 2001; Ortega, 2010; Welter et al., 2013)

 

Sample

The list of companies that operate in the wine sector in the Cava and Rioja PDOs has been compiled from two sources of information. The first source is the database of the Iberian Balance Analysis System (SABI, 2017), taking those companies that are registered and active, under hea-ding 11.02 of the National Classification of Economic Acti-vities (CNAE) corresponding to “Viticultural Companies.” The second source is the public registry of wine companies that exist in the different PDO regions. The number of in-dependent entities that resulted from this compilation was 243 in Cava and 480 in Rioja. Following previous studies, the lost data was eliminated, in this case, eliminating those companies that did not have a valid telephone number or email address (Spanos & Lioukas, 2001). Those companies without a firm structure, or that were a subsidiary of ano-ther company, were also eliminated.

As a result of this process, the total number of indepen-dent companies was reduced to 183 in Cava and 312 in Rioja. A survey was emailed to managers, marketing per-sonnel, or production directors with subsequent telephone reminders a month later. At the end of the process, a total of 66 valid responses were received, 20 in Cava and 46 in Rioja, which represents 13% of the total sample; this has been considered a valid response rate for industrial sectors according to Baruch & Holtom (2008). Table 4 shows the statistical summary of the sample.

 

Variables

The survey was set up after an extensive review of the literature. In addition, to justify its application to the Spa-nish wine sector, a subsequent validation of the survey was carried out among companies, experts, and managers related to the Spanish wine sector. The goal was to ensu-re that the survey was understandable and reflected the peculiarities of the industry. The elements that have been considered are presented below.

 

Managerial capabilities

The managerial capabilities are made up of seven in-dicators, measured by a Likert scale of 5 points where the company had to mark its relative position in front of its competitors from 1, “much weaker than the competitor,” to 5, “much stronger than the competitor.” The ques-tions adapted from Spanos & Lioukas (2001) and Ortega (2010) are: 1) managerial competencies, 2) knowledge and skills of employees, 3) work climate, 4) efficient organizational structure, 5) coordination, 6) strategic planning, and 7) ability to attract creative employees. The survey measured management competence, knowledge and skills of the employees, work environment, efficiency of the organizational structure, coordination, strategic approach, and ability to attract creative employees.

 

Table 4. Statistical summary. Characteristics of the sample.

Source: Author’s own elaboration

 

 

Strategy

To capture the business strategies, the Dess & Davis (1984) and Robinson & Pearce (1988) scales were used to determine the strategic options followed by each winery. See Table S1 [suppl.] for a list of 22 questions asked in our survey. Specifically, each company determined the weight of the four main strategies defined by Robinson & Pearce (1988): efficiency, service, innovation, and marketing. The measure used to capture the strategy consists of a Likert scale with f ive levels where companies rate themselves with respect to different business development efforts, where one is “never used” and five is “main, constantly used.” The main strategies are determined by grouping the 22 questions by strategy.

Managerial capabilities are made up of seven indicators, measured by a five-point Likert scale where the company had to mark its relative position in front of its competitors from one, “much weaker than the competitor,” to f ive, “much stronger than the competitor.” The questions are adapted from Spanos & Lioukas (2001) and Ortega (2010). The items measured were management competence, knowledge and skills of the employees, work environment, efficiency of the organizational structure, coordination, strategic approach, and ability to attract creative employees.

 

Business performance

Following Ortega (2010) and Spanos & Lioukas (2001), the survey have asked managers about business performance in the last three years (period 2013 to 2015, in our case) with seven indicators grouped into two dimensions: market position (sales volume in euros, growth in sales volume in euros, market share in % over sales in euros, and growth in market share over sales in euros), and profitability (profit margin, return on own capital, and net profits). The first dimension shows the external performance of the company, evaluated by its behavior in the market through four items. The second dimension reflects the internal performance of the company, the income generated in its economic activity (Spanos & Lioukas, 2001), through three items. All items use a f ive-point Likert scale, where companies evaluate their position with respect to their competitors, and where the values of the scale are rated from one (“much weaker than the competitor”) to five (“much stronger than the competitor”). Subjective scales are used instead of objective scales, due to two reasons. First, the literature has demonstrated the validity of subjective scales to determine business performance and their convergent validity with objective scales (Dess & Davis, 1984; Richard et al., 2009; Santos & Brito, 2012). Second, accounting data could be subject to annual variability and may include extraordinary results and movements outside the main activity of the company (Richard et al., 2009). Thus, several studies have used subjective instead of objective scales to analyze business performance (Spanos & Lioukas, 2001; Ortega, 2010; Ferrer-Lorenzo et al., 2018; Villanueva & Ferrer, 2020).

 

Methodology

First-step samples comparison between independent

To determine the elements that characterize the PDOs studied, the Mann-Whitney U test was used for two independent samples, Cava and Rioja.

 

Second-step Bayesian regression

Bayesian regressions have been chosen for two reasons. Bayesian regression has greater reliability when the number of cases is low and when the normality of the variables cannot be assured (Block et al., 2011). The proposed analysis model is as follows:

 

 

where the dependent variable Yj is the financial performance value of the company j, measured as the average of the seven items considered in the market performance and financial performance. β0 is the constant; β1, the coefficient of the Robinson and Pearce strategy; β2, the coefficient of managerial capabilities, by extraction of the main components of the seven items analyzed; β3, the Assets of the company, control variable; and finally, ej, the error or the residual of the proposed model. Four regressions (models) are developed that correspond to the four values that the Robinson and Pearce strategy variable can adopt: 1) efficiency strategy, 2) innovation strategy, 3) service strategy, and 4) marketing strategy, as defined in Table S1 [suppl.].. The size of the company has been taken as a control variable, measured on a scale from one to seven, where one represents assets below 400,000 euros, up to seven, representing assets greater than 40 million euros. Table 5 shows the correlation matrix for each of the independent variables. As we can observe, the correlations between some of the strategies proposed in the model have fairly high values, implying serious multicolinearity problems. This high correlation between the strategies prevents their simultaneous inclusion in the models.