Case Study: The Spanish Wine Industry

This scholarly article assesses the elements of competitive advantage in the Spanish wine industry. Strategy, resources, capability, and managerial ability all affect a firm's competitive advantage.

Methodology

Model

In order to determine the relationships between the firm's resources and capabilities, the strategy used, and its performance, this study uses the hierarchical regression method.

In order to adapt the study to the different types of entrepreneurial property that exist in the wine industry in Spain, this study makes three different analyses for the three types of ownership considered: individual companies, cooperatives and mercantile societies. Table 4 reports the distribution of the sample according to the classification made and the wine produced by the companies that answer this question in the questionnaire.

Table 4. Distribution of the different types of ownership studied and the total amount of wine produced by them.

Type of winery	Question: Wine production of the company in hectoliters
Type of winery	Responses	Total hectoliters
Individual Companies	44	44,365
Cooperatives	52	4,629,623
Mercantile Societies	205	1,882,687
Total	301	6,556,675
Spanish wine production in 2015 (OIV, 2016)	37,300,000
Percentage of wine produced by the companies included in the survey over total Spanish wine production	17.6%

Dependent variable

In this paper model the dependent variable is the business performance. This study works under the hypothesis that performance is determined by technological and managerial capabilities, and strategic positioning. The authors have studied the business performance from two approaches, market and finance. , show the answers to the different questions and their frequency.

Table 5. Firm's performance relative to competitor in the last 3 years. Market.

Market		Much Below	Below	Average	Above	Far above	Total
Sales volume €.	n	40	83	129	71	10	333
Sales volume €.	%	12.0	24.9	38.7	21.3	3.0	100.0

Growth in sales volume €.	n	26	58	131	103	15	333
Growth in sales volume €.	%	7.8	17.4	39.3	30.9	4.5	100.0

Market share, % over sales €.	n	41	75	142	67	8	333
Market share, % over sales €.	%	12.3	22.5	42.6	20.1	2.4	100.0

Growth in market share, over sales €.	n	24	59	152	83	13	331
Growth in market share, over sales €.	%	7.3	17.8	45.9	25.1	3.9	100.0

Table 6. Firm's performance relative to competitor in the last 3 years. Financial.

Financial		Much Below	Below	Average	Above	Far above	Total
Profit margin.	n	18	90	147	68	11	334
Profit margin.	%	5.4	26.9	44.0	20.4	3.3	100.0

Return on own capital.	n	28	85	147	62	10	332
Return on own capital.	%	8.4	25.6	44.3	18.7	3.0	100.0

Net profits.	n	23	103	130	67	10	333
Net profits.	%	6.9	30.9	39.0	20.1	3.0	100.0

In order to get a nicely compact representation of the dataset, instead of the original with many variables, this study develops a principal component analysis (PCA), and then the study uses the new component to develop a hierarchical regression.

The principal component analysis is made with the selection of one component, that determines the concept of performance in the firm. The extracted factor explains 66.78% variance, with a KMO = 0.84, Cronbach's alpha = 0.917 (Table 7).

Table 7. Principal component analysis: business performance.

Variables	Alpha without item	Component	Communality
Profitability. Net profits	.902	.836	.698
Market position. Sales volume €	.903	.828	.686
Market position. Market share %	.904	.820	.672
Market position. Growth in market share	.903	.820	.672
Market position. Growth in sales volume €	.905	.813	.661
Profitability. Profit margin	.906	.807	.652
Profitability. Return on own capital	.908	.796	.634
Cronbach alpha of the whole scale	.917
% Total explained variance	66.783
K.M.O.	.840
Barlett Test:	x²	2020.509
	gl	21
	sig	0.000

Independent variables

Technological capabilities, management capabilities, and competitive strategies are set as independent variables.

Technological capabilities

The four indicators used for technological capabilities, their distribution and values are shown in Table 8.

Table 8. Responses and frequency: technological capabilities.

		Much weaker	Weaker	Equal	Stronger	Much stronger	Total
Technological capabilities and equipment	n	44	75	128	73	17	337
Technological capabilities and equipment	%	13.1	22.3	38.0	21.7	5.0	100.0

Efficient and effective production department	n	18	72	139	91	16	336
Efficient and effective production department	%	5.4	21.4	41.4	27.1	4.8	100.0

Economies of scale	n	59	105	100	59	12	335
Economies of scale	%	17.6	31.3	29.9	17.6	3.6	100.0

Technical experience	n	12	43	127	122	32	336
Technical experience	%	3.6	12.8	37.8	36.3	9.5	100.0

In order to introduce the variable in the linear regression model and to avoid multicollinearity, this paper performed the technique of principal component analysis. One extracted factor explains 57.9% of variance, KMO = 0.71, and Cronbach's alpha = 0.751 (Table 9). The component has been called "technological capability".

Table 9. Principal component analysis: technological capabilities.

Variables	Alpha without item	Component	Communality
Efficient and effective production department.	.613	.864	.746
Technological capabilities and equipment.	.709	.741	.549
Economies of scales.	.715	.725	.525
Technical experience.	.728	.704	.496
Cronbach alpha of the whole scale	.751
% Total explained variance	57.914
K.M.O.	.713
Barlett Test:	x²	339.887
	gl	6
	sig	.000

Managerial capabilities

The seven indicators used for managerial capabilities, their distribution and values are shown in Table 10.

Table 10. Responses and frequency: managerial capabilities.

		Much weaker	Weaker	Equal	Stronger	Much stronger	Total
Managerial competencies.	n	11	48	188	73	16	336
Managerial competencies.	%	3.3	14.3	56.0	21.7	4.8	100.0

Knowledge and skills of employees.	n	6	23	172	108	24	333
Knowledge and skills of employees.	%	1.8	6.9	51.7	32.4	7.2	100.0

Work climate.	n	6	9	133	142	41	331
Work climate.	%	1.8	2.7	40.2	42.9	12.4	100.0

Efficient organizational structure.	n	9	30	177	95	21	332
Efficient organizational structure.	%	2.7	9.0	53.3	28.6	6.3	100.0
Coordination.	n	9	31	167	106	19	332

The seven indicators have been reduced following the principal component analysis. Then the new component has been used to develop a hierarchical regression. Resulting one factor that explains 61.6% of the variance with KMO = 0.87, and Cronbach's alpha = 0.895 (Table 11). The component is called "managerial capability".

Table 11. Principal component analysis: managerial capabilities.

Variables	Alpha without item	Component	Communality
Strategic planning	.873	.832	.692
Efficient organizational structure.	.875	.824	.678
Coordination.	.876	.818	.669
Ability to attract creative employees.	.883	.773	.597
Work climate.	.883	.766	.586
Knowledge and skills of employees.	.882	.765	.585
Managerial competencies.	.889	.717	.514
Cronbach alpha of the whole scale	.895
% Total explained variance	61.650
K.M.O.	.870
Barlett Test:	x²	1243.602
	gl	21
	sig	.000

Business strategy

In order to manage the main indicators of the firm's strategy, principal component analysis has been used. In Table 12 the components obtained from the analysis can be seen.

Table 12. Principal component analysis: strategy of the firm.

Variables	Alpha without item	Comp. 1	Comp. 2	Comp. 3	Comp. 4	Comp. 5	Communality
Extremely strict product quality control procedures.	.870	.704	.059	.100	-.243	.136	.587
Specific efforts to insure a pool of highly trained, experienced personnel.	.866	.665	.278	.207	-.023	-.005	.562
Continuing, overriding concern for lowest cost per unit.	.871	.649	.062	.132	.323	-.048	.549
Major effort to ensure availability of raw materials.	.870	.643	.254	-.071	.113	.025	.496
Extensive customer service capabilities.	.871	.565	.015	.368	-.149	-.043	.479
Maintaining high inventory levels (disregard the derivative of the aging of the product).	.870	.535	.189	.007	.250	.260	.452
Concerted effort to build reputation within industry.	.865	.518	.240	.384	-.269	.293	.632
Building brand identification.	.867	.489	.400	.236	-.233	.106	.521
Developing and refining existing products.	.867	.474	.207	.322	-.210	.306	.510
Promotion/advertising expenditures above the industry average.	.869	-.012	.826	.148	.158	.043	.732
Major expenditure on production process oriented R&D.	.865	.281	.766	.063	.092	.130	.695
Innovation in marketing techniques and methods.	.866	.204	.742	.226	-.058	.015	.647
Strong influence over distribution channels.	.865	.299	.659	.223	.129	.057	.593
Innovation in manufacturing process.	.864	.385	.443	.341	.005	.253	.525
New product development.	.868	.164	.241	.790	.127	-.093	.733
Broad product range.	.870	.207	.240	.727	.262	-.273	.772
Emphasis on the manufacturing of speciality products.	.869	.139	.200	.680	-.209	.247	.627
Products in higher priced market segments.	.872	.143	.196	.471	-.438	.404	.635
Pricing below competitors.	.882	-.075	.105	.060	.796	.129	.670
Products in lower priced market segments.	.879	.072	.125	-.023	.786	.086	.647
Small limited range of products.	.879	.203	.027	-.249	.056	.773	.705
Only serve specific geographic markets.	.876	-.009	.106	.158	.177	.715	.579
Eigen value		6.767	2.275	1.783	1.419	1.103
% Explained variance		30.758	10.339	8.107	6.448	5.013
Cronbach's alpha of whole scale:		.875
% Total explained variance		60.663
Average K.M.O.		.862
Bartlett Test
x²		2557.814
gl		231
Significance		0.000

In this case, five components have been extracted, 1) Efficiency, 2) Marketing, 3) Innovation, 4) Low Price, and 5) Small Market and Product. The set explains 60.66% of the variance. The results of the different statistical reliability have values within the limits of acceptability, Cronbach's alpha = 0.875 and KMO = 0.862.

Efficiency strategy: Nine issues out of the twenty defined by Robinson and Pearce, are part of this first extracted component, accounting for 30.8% of the variance. Efficiency Strategy contains concepts that lead the company to the extreme care of the products offered to the customer and ensure the realization of an efficient process including: strict quality control, trained and experienced staff, encourage available raw materials, improve cost per unit, high level of inventory, customer service, promote reputation in the industry, brand identification and development of existing products. It is important to state that in this sector, the high level of inventory is relevant, as the development of aged products through aging and reserves generate higher added value.

Marketing strategy: In this second component five questions explain 10.3% of the variance. They are: advertising spending above sector average, investment in R&D oriented to efficiency process, innovations in marketing, strong influence over distribution and innovation in productive process. In this area business managers are concerned about trends and about controlling their various marketing techniques as a strategy to achieving their success.

Innovation strategy: The variance explained by the extracted component is 8.1% and its four questions are: development of new products, wide range of products, emphasis on special products and high-price segment. In innovation strategy what prevails is the obtaining of new items and the ability to offer the market a new and special range of products with a certain orientation towards a greater perceived benefit by customers.

Low price strategy: Two variables characterize this factor and explain 6.4% of the variance they point in a clear direction for offering products with less perceived benefit, a price below competitors and focusing on the low price products segment.

Small market and product strategy: This component refers to those companies that choose to compete through a strategy of limited or specialized products, more oriented to high rather than low prices and to a very specific segment. The total variance explained in this case is 5.0%.

Cost strategy and differentiation strategy. Porter's generic strategies: In the design of the hypothesis for the theoretical analysis of the strategic options, authors have adopted Porter's model (1980 and 1985), and its two generic strategies: cost or differentiation. However, our factor analysis reveals five different strategies: efficiency, marketing, innovation, low prices, small market and product strategy. The efficiency strategy refers to the maximum control of resources and is part of Porter's overall strategy of costs. The innovation strategy contemplates offering better products with greater added value to the customers and is part of the differentiation strategy. The marketing strategy is a strategy that is used by both generic strategies and contemplates the sensitivity to the market and its adaptation to changes, brand image and control of distribution. The low-price strategy has been assigned to a cost strategy and a small market and product strategy, which refer to Porter's third focus strategy. Table 13 shows the relation between the extracted strategies and Porter's generic strategies.

Table 13. Relation between extracted strategies and Porter's strategies.

Strategies extracted	Porter's generic strategies
Efficiency Strategy	Cost Strategy
Low Price Strategy	Cost Strategy
Innovation Strategy	Differentiation Strategy
Small Market and Product Strategy	Focus Strategy
Marketing Strategy	No assignation, the study maintains as Marketing Strategy

Control variables

The purpose of the study is to determine the relationship between business performance and the set of independent variables that have been defined. However, numerous studies refer to the influence on performance of elements such as company size and the degree of rivalry. In the Italian wine industry Sellers and Alampi-Sottini have found a positive correlation with the influence of size in a winery's performance by studying the size of the company, with the number of employees, total turnover and volume of assets. Therefore, most of the studies include control variables that help to better understand the business result. In this study authors will take the size of the company and level of rivalry within the sector as control variables. The size of the company will be measured in terms of its assets, divided into seven categories, ranging from less than 400 thousand euros to more than 20 million euros. Other studies take the number of employees or its logarithm. In this case, authors have opted for assets due to the extremely seasonal nature of wine production and the resulting distortion figures produced. Table 14 reports the distribution of the assets in the sample.

Table 14. Distribution of the different types of ownership studied and volume of assets.

Type of winery studied and volume of assets in millions of euros.	Individual companies		Cooperatives		Mercantile companies		Total
	n	%	n	%	n	%	n	%
< 0.4 million	30	63.8%	6	11.5%	62	29.8%	98	31.9%
0.4 - 1 million	6	12.8%	12	23.1%	51	24.5%	69	22.5%
1-5 million	11	23.4%	21	40.4%	65	31.3%	97	31.6%
5-10 million	0	0.0%	10	19.2%	15	7.2%	25	8.1%
10-20 million	0	0.0%	0	0.0%	8	3.8%	8	2.6%
20-40 million	0	0.0%	2	3.8%	4	1.9%	6	2.0%
> 40 million	0	0.0%	1	1.9%	3	1.4%	4	1.3%
Total	47	100.0%	52	100.0%	208	100.0%	307	100.0%
No responses							32

The degree of intensity of rivalry has been measured with the scale used by Ortega and Spanos and Lioukas. The manager of the firm evaluates the level of competition with a 5-point Likert scale which evaluates the characteristics of the product, promotion strategies, access to distribution channels and customer service strategy. The extracted factor explains 69.2% of variance, KMO=0.80, and Cronbach's alpha=0.85 (Table 15). The component has been called "internal rivalry".

Table 15. Principal component analysis: internal rivalry.

Variables	Alpha with out item	Component	Communality
Promotional strategies among rivals.	.785	.874	.768
Service strategies to customers.	.808	.843	.711
Access to distribution channels.	.812	.839	.709
Product characteristics.	.846	.769	.599
Cronbach alpha of the whole scale	.850
% Total explained variance	69.266
K.M.O.	.805
Barlett Test:	Chi-squared	574.787
	gl	6
	sig	.000

Proposed model

The proposed model of analysis is as follows:

$Y_j = β_0 + β_1Cr_j + β_2Ca_j + β_3Ee_j + β_4Em_j + β_5Ep_j + β_6Eb_j + β_7Es_j + β_8Rt_j + β_9Rm_j + e_i$

Where,

$Y_j$ is the performance value for the company "

$j$ ";

$β_0$ is the constant;

$β_1$ , and

$β_2$ , the coefficients of the control variables: internal rivalry and assets;

$β_3$ ,

$β_4$ ,

$β_5$ ,

$β_6$ ,

$β_7$ , the coefficients of the variables efficiency, marketing, innovation, low price and small market and product.

$β_8$ and

$β_9$ , are the coefficients for the resources and capabilities of the firm: technology and managerial capabilities. And ej is the error or the residual of the proposed model.

The variables chosen to build the multivariable linear correlation, the mean values, standard deviation, Cronbach's alpha without item and correlation matrix, are shown in Table 16. The Cronbach alpha of the whole scale is 0.635. The study evaluates the possible multicollinearity between the variables through FIV and conditioning index, in both cases the values are lower than ten, as recommended by the literature.

Table 16. Correlation matrix.

	Mean	SD	Cronbach's Alpha Without Item	1	2	3	4	5	6	7	8	9	10
(1) Internal Rivalry	0	1.00	.637	1
(2) Assets	2.38	1.30	.582	.100	1
(3) Efficiency Strategy	0	1.00	.620	.190^**	.133^*	1
(4) Marketing Strategy	0	1.00	.601	.072	.300^**	.000	1
(5) Innovation Strategy	0	1.00	.640	.143^*	.133^*	.000	.000	1
(6) Low Price Strategy	0	1.00	.655	-.087	.187^**	.000	.000	.000	1
(7) Small Market and Product Strategy	0	1.00	.671	.017	-.026	.000	.000	.000	.000	1
(8) Technological Capability	0	1.00	.541	.120^*	.434^**	.306^**	.448^**	.074	-.014	.036	1
(9) Managerial Capability	0	1.00	.572	.164^**	.204^**	.308^**	.347^**	.185^**	-.148^*	.103	.482^**	1
(10) Performance	0	1.00	.549	.037	.360^**	.251^**	.449^**	.226^**	.048	.025	.550^**	.412^**	1

** The correlation is significant at the 0.01 level (bilateral).
* The correlation is significant at the 0.05 level (bilateral).

The model has been run with SSPS v20 program with the introduction hierarchical method.