Wine Economics and Policy 12(1): 37-49, 2023 Firenze University Press www.fupress.com/wep ISSN 2212-9774 (online) | ISSN 2213-3968 (print) | DOI: 10.36253/wep-11522 Wine Economics and Policy Citation: Kamila Vesela, David Křížek, Lucie Severova (2023). Structure and development of the Czech wine market and foreign wine trade. Wine Economics and Policy 12(1): 37-49. doi: 10.36253/wep-11522 Copyright: © 2023 Kamila Vesela, David Křížek, Lucie Severova. This is an open access, peer-reviewed article published by Firenze University Press (http://www.fupress.com/wep) and dis- tributed under the terms of the Crea- tive Commons Attribution License, which permits unrestricted use, distri- bution, and reproduction in any medi- um, provided the original author and source are credited. Data Availability Statement: All rel- evant data are within the paper and its Supporting Information fi les. Competing Interests: The Author(s) declare(s) no confl ict of interest. Structure and development of the Czech wine market and foreign wine trade Kamila Vesela, David Křížek*, Lucie Severova Czech University of Life Sciences Prague, Faculty of Economics and Management, Kamýcká 129, 165 00 Prague, Czech Republic E-mail: veselakamila@pef.czu.cz; krizekd@pef.czu.cz; severova@pef.czu.cz *Corresponding author. Abstract. Th e aim of the presented text is to evaluate the structure and development of the Czech wine market, foreign wine trade and to analyse the factors shaping domestic demand for wine. Th e text presents the results of the analysis of primary and second- ary data. Time series on the development and direction of the foreign wine trade are analysed and the results are then compared with the fi ndings of a questionnaire survey of 946 respondents which provides information about the current consumer prefer- ences of the Czech population in drinking wine. Wine consumption has shown sta- ble growth over the last 20 years, as demonstrated by primary research. Th e research has proven that wine consumption in the Czech Republic shows a faster growth rate than domestic production, which is also refl ected in import demand. From the point of view of factors infl uencing domestic demand for wine, it is mainly the age, education, income and size of the consumer’s residence. Th e research provides an up-to-date view of the structure of wine demand in the Czech Republic and identifi es the factors infl u- encing wine demand. Th e research also makes it possible to predict the future direc- tion of the Czech wine trade. Keywords: export, import, production, consumption, wine. 1. INTRODUCTION Th e Czech Republic is a country with long-term wine production, but also consumption. From 1995 to 2016, wine production increased from 459 (in 1000 hl) to 631 (1000 hl). Th is production grew at an average annual rate of 4.57% per year. In contrast, wine consumption in the Czech Republic increased from 1995, when its value was 63.7 (1000 hl) to 192 (1000 hl). It is an increase of more than 300%. Th e average consumption per person in 1995 was 7.6 l of wine, in 2016 it was already 21.3 l/person. Th e average growth rate of wine consumption was 6.57% per year. For a clear overview, see fi gure 1. A similar trend can be observed in Asia, where the share of consumer spending in total household spending is also growing. Th e largest consump- tion of wine is currently in France, followed by Italy and Austria. Bentzen and Smith [1] dealt with the situation and problems of wine production in 38 Kamila Vesela, David Křížek, Lucie Severova countries with low wine production, especially in Den- mark, whereas Marquart and Hanf [2] focused on Arme- nia. Wine production in the Czech Republic is more focused on white wine. The share of white grape varie- ties in young vineyards in the Czech Republic in 2015– 2018 was 92% of the total area of vineyards. Veltlínské zelené, Pálava and Ryzlink rýnský [3] have the largest share in the Czech Republic. The unequivocal market leader is the Bohemia sekt group [4]. Chládková [5] adds sufficient investment in the development and expan- sion of market share as a key element for maintaining the company’s competitive position in the wine market. It should also be noted that the Czech Republic’s acces- sion to the European Union significantly affected the wine market in the Czech Republic, which was reflect- ed, among other things, in a slight decline in the mar- ket shares of most companies [6]. For example, Butkus et al. [7] dealt with the issue of the impact of the Czech Republic’s accession to the European Union on the Czech Republic’s foreign trade. They used an econo- metric model to try to determine the direct effect of the Czech Republic’s accession to the EU on the size of for- eign trade. However, the analyses carried out provided mixed results regarding EU membership and export growth. With the accession of the Czech Republic to the European Union, there have been a number of changes in Czech viticulture, the market for Czech consumers now offers a number of foreign wines in various price categories and quality levels. The integration of domes- tic and foreign wine markets can be expected, which will be reflected in the gradual convergence of wine prices at the global level. Most Czech producers are convinced that their wines will continue to be sufficiently competi- tive [8]. The development of the wine industry contrib- utes to the overall competitiveness of individual regions of the Czech Republic, as well as the competitiveness of the Czech Republic on global markets [9]. Specific prob- lems of this sector including the development of wine consumption per capita were analysed, for example, by Chládková [10]. Verner [11] analysed the relationship between economic growth, production growth in the sector and quality of life. To understand the functioning of a particular market, in addition to knowledge of sec- ondary data on overall market variables (such as produc- tion, consumption, profitability, or foreign trade), it is important to know the factors influencing the behaviour of individual market players. The importance of studies of agricultural commodities in the Czech Republic was emphasized e.g. by Svatoš and Smutka [12]. Wine is one of the commodities with a long history of production and consumption in the Czech Repub- lic. The aim of this text is to evaluate the structure and development of the Czech wine market, foreign trade in wine and to analyse the factors shaping domestic demand for wine. The research provides an up-to-date view of the structure of wine demand in the Czech Republic and identifies the factors inf luencing wine demand. The research also allows predicting the future direction of the Czech wine trade. This is important for regional development planning, as wine production is significantly more widespread in certain regions than in others (due to climatic conditions). The aim of the presented text is a comprehensive evaluation of the Czech wine market in terms of wine production, consumption and foreign trade. The fol- lowing sub-objectives have been set, the fulfilment of which will lead to the fulfilment of the main objective: 1) to evaluate the development and predict production, consumption and foreign trade (exports and imports) in wine in the Czech Republic; 2) to analyse the factors of wine demand among the population of the Czech Republic. 2. MATERIAL AND METHODS The key source of data for this research are 1) sec- ondary data coming from the official website of the Czech Statistical Office (hereinafter referred to as the CSU) and the Ministry of Agriculture of the Czech Republic (hereinafter referred to as the MZCR); 2) pri- mary data obtained by a questionnaire survey. The data base of secondary data consists mainly of annual data on the development of wine production and consumption in litres, as well as wine exports and imports in litres and average wine consumption per person (in litres/ year). In the research of secondary data, both absolute and relative indicators are analysed. Secondary data for analysis are available in a comprehensive version for the 0 500 1000 1500 2000 2500 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 10 00 h l Wine production Wine consumption Figure 1. The development of wine production and consumption in the Czech Republic over the period 1995-2016 (Source: own pro- cessing based on MZCR data). 39Structure and development of the Czech wine market and foreign wine trade period 1995–2016 (unfortunately, newer data are not yet available). For the basic evaluation of time series, basic statisti- cal descriptive characteristics were used, which were uti- lized to describe the year-on-year rates and the absolute deviation and their development between wine produc- tion and consumption and wine exports and imports. Methods of trend analysis and regression statistics were used for analysis and prediction. The selection of a suit- able trend function was made by using and evaluating interpolation criteria. The function chosen is that which has the smallest possible value of the MSE criterion, meets the conditions of the F-test and has a sufficiently high value of reliability R. The value of MSE is deter- mined using function (1): (1) where T is the number of observations, yt are measured values and are expected values. Linear and quadratic curves were used as basic with respect to the develop- ment of time series. The following null hypotheses were established: - H01: The wine production/consumption in the Czech Republic has no trend (is stationary); - H02: The development of the wine export/import ratio in the Czech Republic has no trend. (is stationary). The results of the analysis of time series are further confronted with the results of the questionnaire sur- vey, which was conducted in 2020 and was focused on the evaluation of consumer preferences of the popula- tion of the Czech Republic regarding wine consump- tion. In order to obtain up-to-date consumer informa- tion, primary research was carried out to determine the extent, size and structure of the demand for wine. The aim was in particular to identify the factors influencing the size and formation of demand for wine. Quantitative research using a questionnaire technique of data collec- tion was used to obtain primary data. The research was carried out in 2020 in September and October. A total of 946 respondents took part in the questionnaire survey, which represented an 84% return after the elimination of empty or incomplete questionnaires. The gender com- position of the respondents was 420 (44.4%) men and 526 (55.6%) women. The age structure of the respond- ents was divided into groups of 0–20 years, 21–25 years, 26–30 years, 31–40 years, 41–50 years, 51–60 years and 61 and more years. The dominant groups are 0–20 years and 21–25 years, which together make up 86% of all respondents. The structure of respondents’ demand was further examined according to the highest level of education attained and the amount of income. Of the total number of respondents, 81.1% of respondents indi- cated that they would indulge in wine at least occasion- ally. Absolute and relative frequencies were used in the descriptive statistics and contingency tables and the χ2 test was used to analyse the obtained data. The con- tingency table contains the observed frequencies from the questionnaire survey of individual combinations of characters. From the differences (residues) of the observed frequencies and the frequencies obtained from the assumption of the null hypothesis, the total normal- ized residue is calculated. If its value is less than the crit- ical value of the distribution χ2 at the significance level of 0.95 for the appropriate degree of freedom, the null hypothesis cannot be rejected at the 95% significance level. Significant variables (according to Chi-square test) were further tested using multinomial and ordinal logistic regression, depending on the type of dependent variable. Logistic regression models were constructed and their quality was evaluated based on Nagelkerke‘s R squared value. Furthermore, the significance of the regressors was tested using Omnibus Likelihood Ratio Tests. The questionnaire survey focused on the following questions: How often do you drink wine? What wine do you prefer? Where do you most often buy wine? In what price range do you most often buy wine? The answers to these questions were then analysed according to gender, age, education attained, income and size of residence of the respondents. For these purposes, null hypotheses were established and tested. 3. RESULTS The first phase of the research is focused on evalu- ating the state of the Czech wine market and the pos- sibilities of predicting further market direction. For these purposes, time series on the production and con- sumption of wine in the Czech Republic and the values of exports and imports of wine from or to the Czech Republic are analysed. In the next phase, the results of this research are compared with the results of the analy- sis of primary data from the questionnaire survey. The following graph (Figure 2) maps the develop- ment of wine production and consumption in the Czech Republic for the period 1995–2016. Wine production shows grow th in the period observed, but it also shows relatively high variations. This phenomenon must be attributed to the fact that wine production is conditioned by a number of factors. First of all, it is the area of vineyards, then especially 40 Kamila Vesela, David Křížek, Lucie Severova the quality and extent of the harvest in individual years, which is conditioned by the quality of climatic condi- tions, the number of pests and other factors. In the whole period observed, wine production exceeds con- sumption, however, the absolute difference between pro- duction and consumption decreases over time. This is due to the faster growth rate of wine consumption over production. This effect is clearly evident from the indica- tor of wine production/consumption where, as the graph above shows, this indicator has been declining for a long time. It is this indicator of the ratio of the amount of production and consumption of wine that expresses the direction of the Czech wine market and was the subject of analysis and prediction. For these purposes, a null hypothesis was established: - H01: The ratio of the amount of production and consump- tion of wine in the Czech Republic has no trend. The relationship between wine production and con- sumption was expressed using a ratio indicator found as the ratio of total wine production in the Czech Republic/ total wine consumption in the Czech Republic. As the graph above shows, this indicator has a declining trend, indicating a faster increase in wine consumption com- pared to production. Subsequently, a regression analysis was performed in order to reveal and describe the func- tion characterizing the development of the share of wine production and consumption. Using the quadratic trend function, it was possible to explain 74.34% of the vari- ability of the dependent variable; the value of the F test, or the P value shows a value of 0.000002, i.e. it satisfies the condition of a result with less than 5% level of sig- nificance. The individual parameters of the function are described in the following table (Table 1). The functional relationship is described by equation (2): yt = 9,1473-0,5587t + 0,0131t2 (2) where yt is the ratio of wine production and consump- tion in the Czech Republic in individual years. Using this function, the values for the expected development of production and consumption for the next 4 periods are then simulated. These are shown in the following graph (Figure 3). The research has shown a relationship between wine production and consumption, which can be described by this function. The development showed a faster growing rate of wine consumption than production, which reduc- es the difference between production and consump- tion, and thus also decreases the analysed ratio indica- tor of wine production/consumption. This is followed by research into whether and how this growing con- sumption is reflected in the foreign trade of the Czech Republic. For these purposes, time series on the export and import of wines from and to the Czech Republic are analysed. Data in this case are available until 2019. The development of the foreign wine trade is analysed in summary and according to the division into vermouth, sparkling wine and grape wine (still). For these purpos- es, a null hypothesis was established: - H02: The development of the wine export/import ratio in the Czech Republic has no trend. The relationship between wine exports and imports was expressed using a ratio indicator found as the ratio of wine exports from the Czech Republic/wine imports to the Czech Republic. The following graph (Figure 4) describes the development of the foreign wine trade. 0 1 2 3 4 5 6 7 8 9 10 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 1… 1… 1… 1… 1… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… 2… Co ns um pt io n/ pr od uc tio n 10 00 h l Figure 2. The development of wine production and consumption in the Czech Republic over the period 1995-2016. Blue columns indi- cate wine production. Orange shows wine consumption. The grey column is wine production minus consumption. The yellow curve shows the ratio of production to consumption (Source: own pro- cessing based on MZCR data). Table 1. Parameters of the function of the ratio of production and consumption of wine in the Czech Republic. Reliability value R SS residues MSE F test (“P”) parameter a parameter b parameter c 0.74343514 21.93117 0.996871 0.000002 9.147321 -0.55866 0.013096 Source: own processing. 41Structure and development of the Czech wine market and foreign wine trade Figure 4 shows the growth of wine imports, which corresponds to the growing demand, or wine consump- tion. The import of wine in 2019 increased almost three- fold compared to 1999, i.e. approximately the same as the consumption of wine in the Czech Republic. Tomšík [13, 14] adds that the largest volumes of wine are imported from Italy and Hungary. The export of wine then goes mainly to Slovakia and Poland. In the case of Slovakia, the export of wine from the Slovak Republic is directed, among other destinations, to the Czech Repub- lic [15]. Subsequently, a regression analysis was performed in order to reveal and describe the function character- izing the development of the share of wine exports and imports. In the case of total exports and imports, it is not possible to describe the development of the share of exports and imports of wine using a suitable function and it must be stated that there is no statistically signifi- cant relationship describing the development of exports and imports of wine. A more detailed analysis according to the type of wine has already been able to describe a statistically significant relationship, broken down into foreign trade in vermouth, sparkling wine and grape wine (still). The development of foreign trade in ver- mouth is shown in the following graph (Figure 5). Dur- ing the period observed, imports of vermouth increased by 2.7 times the value of 1999, while exports fell to about one third. Using the quadratic trend function, it was possi- ble to explain 55.90% of the variability of the depend- ent variable of the ratio of exports and imports, yet the model shows statistical significance; the value of the F test, or the P value shows a value of 0.0006, thus satisfy- ing the condition of a result with less than a 5% level of significance. The individual parameters of the function are described in the following table (Table 2). The functional relationship is described by equation (3): yt = 1,0097 – 0,1294t + 0,0045t2 (3) where yt is the ratio of vermouth export and import over time. Using this function, the values for the expected development of foreign trade in vermouth for the follow- ing period are then simulated. These are shown in the following graph (Figure 6). 0 2 4 6 8 10 12 14 16 18 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 Pr od /c on s. Figure 3. Prediction of the development of wine production/con- sumption in the Czech Republic. The orange curve shows an esti- mate. The blue curve represents production and consumption (Source: own processing). 0,00000000 0,05000000 0,10000000 0,15000000 0,20000000 0,25000000 -200.000.000 -150.000.000 -100.000.000 -50.000.000 0 50.000.000 100.000.000 150.000.000 200.000.000 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 Ex po rt /i m po rt N et to k g Figure 4. The Czech Republic’s foreign wine trade. The blue column indicates import. Orange indicates export. Gray shows net exports. The yellow curve represents the export/import (Source: own pro- cessing based on CSU data). 0 0,2 0,4 0,6 0,8 1 1,2 1,4 -8.000.000 -6.000.000 -4.000.000 -2.000.000 0 2.000.000 4.000.000 6.000.000 8.000.000 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 Ex po rt /I m po rt N et to k g Figure 5. The Czech Republic’s foreign trade in vermouth. The blue column indicates import. Orange indicates export. Gray shows net exports The yellow curve shows exports/imports (Source: own pro- cessing based on CSU data). Table 2. Parameters of the function of the ratio of export and import with vermouth. Reliability value R SS residues MSE F test (“P”) parameter a parameter b parameter c 0.559041 0.943614 0.044934 0.00063 1.009714 -0.12944609 0.00446645 Source: own processing. 42 Kamila Vesela, David Křížek, Lucie Severova In the case of sparkling wine, there was an even greater increase in imports. Compared to 1999, the import of sparkling wine increased more than 6-fold by 2019, while exports, similarly to vermouth, decreased in the period observed (Figure 7). Using the quadratic trend function, it was possi- ble to explain 73.88% of the variability of the depend- ent variable of the ratio of exports and imports, yet the model shows statistical significance; the value of the F test, or the P value shows a value of 0.00001, thus satis- fying the condition of a result with less than a 5% level of significance. The individual parameters of the func- tion are described in the following table (Table 3). The functional relationship is described by equation (4): yt = 0,6465 – 0,0584t + 0,0014t2 (4) where yt is the ratio of export to import of sparkling wine over time. Using this function, the values for the expected development of foreign trade in sparkling wine for the following period are simulated. These are shown in the following graph (Figure 8). Unlike vermouth and sparkling wine, still grape wine showed significant growth on the export side during the period observed. Compared to 1999, wine exports in 2019 reached more than 8 times the value in 1999. At the same time, imports in this case increased “only” 2.7 times over the period observed (Figure 9). The development of the foreign trade relationship, or the mutual ratio of exports and imports is more com- plicated in this case. The quadratic trend function was able to explain 42.33% of the variability of the depend- ent variable, yet the model shows statistical significance. The individual parameters of the function are described in the following table (Table 4). 0 0,2 0,4 0,6 0,8 1 1,2 1,4 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 20 21 20 22 Ex po rt /i m po rt Figure 6. Predicting the development of foreign trade in vermouth. The blue curve indicates export/import and the orange estimate (Source: own processing). 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 -10.000.000 -8.000.000 -6.000.000 -4.000.000 -2.000.000 0 2.000.000 4.000.000 6.000.000 8.000.000 10.000.000 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 Ex po rt /I m po rt N et to k g Figure 7. The Czech Republic’s foreign trade in sparkling wine. The blue column indicates import. Orange indicates export. Gray shows net exports. The yellow curve represents the export/import (Source: own processing based on CSU data). Table 3. Parameters of the function of the ratio of export and import with vermouth. Reliability value R SS residues MSE F test (“P”) parameter a parameter b parameter c 0.738835 0.213037 0.010145 0.00001 0.646529 -0.05843594 0.00143415 Source: own processing. 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 20 21 20 22 Ex po rt /I m po rt Figure 8. Predicting the development of foreign trade in sparkling wine. The orange curve shows an estimate. The blue curve repre- sents production and consumption (Source: own processing). 0 0,05 0,1 0,15 0,2 0,25 -200.000.000 -150.000.000 -100.000.000 -50.000.000 0 50.000.000 100.000.000 150.000.000 200.000.000 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 Ex po rt /I m po rt N et to K g Figure 9. The Czech Republic’s foreign trade in grape wine – still. The blue column indicates import. Orange indicates export. Gray shows net exports. The yellow curve represents the export/import (Source: own processing based on CSU data). 43Structure and development of the Czech wine market and foreign wine trade The functional relationship is described by equation (5): yt = -0,0462 + 0,0229t – 0,0008*t2 (5) where yt is the ratio of export to import of still wine over time. Using this function, the values for the expected development of foreign trade in still wine for the follow- ing period are simulated. These are shown in the follow- ing graph (Figure 10). The conclusions from the first phase of the research are as follows: Although the volume of wine production in the Czech Republic outweighs its consumption, in the long run and in the whole period observed there is a faster growth rate of wine consumption than produc- tion. Thus, the ratio indicator of production/consumption decreases; this phenomenon proved to be statistically sig- nificant. The growing consumption of wine was also evi- denced by the growing demand for wine imports, which also showed a total of about 3-fold growth between 1999 and 2019 (i.e., approximately the same growth as the growth of total wine consumption in the Czech Repub- lic in the given period). Imports of sparkling wine are growing the fastest (6 times the imports between 1999 and 2019), vermouth and still wine from grapes show an increase of 2.7 times over the period observed from 1999. In terms of volume, however, the largest item is the import of still wine. In the case of still wine from grapes, it was the only item mentioned for which exports increased, particularly by 8 times compared to 1999. The trends observed in exports and imports for individual wines proved to be statistically significant in all cases. The second part of the research is based on the evaluation of primary data obtained from a question- naire survey. The aim of this research was mainly to test the results of previous research resulting from the analysis of secondary data. Given the findings of grow- ing wine consumption, respondents were first asked if they drank wine and how often. Respondents chose from the answers: daily, every other day, once a week, once a month and exceptionally. The highest frequency was recorded in the answer “exceptionally” (383 answers) and “once a week” (263 answers), the lowest frequency, on the other hand, was recorded in the answer “daily” (10 answers). Furthermore, factors that may affect the intensity of wine drinking were analysed, namely gen- der, age, education, income and size of residence of the respondents. In this context, the following null hypoth- esis was established: - H03: The gender, age, education, income and size of resi- dence do not significantly predict the frequency of wine drinking. Multinomial and ordinal logistic regression were used to evaluate the statistical hypothesis. First, ini- tial preparation for the use of logistic regressions was performed. A separate table of results is created for each dependent variable to determine the relationship between the dependent and independent variables. The dependencies were examined using Chi-square test. The significant variables were then entered into an ordinal logistic regression model. The results of the first hypothesis testing by the chi- square test are shown in the following Table 5. Table 4. Parameters of the function of the ratio of export and import with still wine from grapes. Reliability value R SS residues MSE F test (“P”) parameter a parameter b parameter c 0.423322 0.041582 0.00198 0.007053 -0.04624 0.02292227 -0.0008445 Source: own processing. -0,05 0 0,05 0,1 0,15 0,2 0,25 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 20 21 20 22 Ex po rt /I m po rt Figure 10. Predicting the development of foreign trade in still wine. The orange curve shows an estimate. The blue curve represents pro- duction and consumption (Source: own processing). Table 5. The results of Chi-square test of the question “How often do you drink wine?”. Criterion χ2 Critical value Result gender 32.6112 9.487729 rejected age 16.38039 15.50731 rejected education 25.39251 9.487729 rejected income 19.84832 21.02607 not rejected size of residence 23.76653 26.29623 not rejected Source: own processing. 44 Kamila Vesela, David Křížek, Lucie Severova The research has shown that there is a relationship between gender and the frequency of wine drinking. Based on the results of the Chi-square test, an Ordi- nal Logistic Regression model was created. The refer- ence category was the last and most numerous category “exceptionally”. Only variables that were statistically sig- nificant according to the Chi-square test were included in the model. The results of the ordinal logistic regression are shown in the following Table 6. Based on the result of the Likelihood Ratio test, we accept the alternative hypothesis, i.e. at least one predic- tor is statistically significant at the 5 % significance lev- el. Nagelkerk’s pseudo R2 of 0.02 indicates a very small effect of the predictors on the explanatory variable. Table 7 summarizes the impact of each predictor. It is clear from the table that at the standard 5 % signifi- cance level, the variables education and gender are influ- ential in the model. The answers are in line with the findings of Chlád- ková et al. [16], who showed in her research based on direct questioning of 1,000 respondents from all over the Czech Republic that 11.3% of respondents drink wine several times a week, 29.8% of respondents drink wine at least once a week and 23.32% of respondents drink wine several times a month. She also noted that only 2.7% of respondents do not drink wine at all. Kel- ley et al. [17] also concluded that wine is drunk more by women, but argues that the frequency of consumption is higher in men than in women. An interesting feature of his research is also the research on the importance of the relationship between information about the pairing of food and wine on the labels of bottles and demand, or consumption of wine. As a result, consumers who buy wine at least once a week were positively affected by this information, and this effect decreased with the frequen- cy of purchase. The second research question was: “What wine do you prefer? White, rosé or red?” White wine had the highest frequency of responses (664 responses), rosé had the lowest frequency of responses (187). Subsequently, null hypothesis was established: - H04: The gender, age, education, income and size of resi- dence do not significantly predict the wine type preference (White, rosé or red). The results of the chi-square test are shown in the following Table 8. The research has shown that there is a relationship between wine type selection and gender, age, education and income. The relationship between the size of resi- dence and wine type selection was not confirmed. Based on the Chi-square test results, a Multinomial Logistic Regression model was created. The reference category was the most numerous category – white wine and only variables that have a significant effect on the dependent variable according to the Chi-square test were included in the model. The results of the multinomial logistic regression are shown in the following table Table 9. Based on the result of the Likelihood Ratio test, we accept the alternative hypothesis, i.e. at least one predic- tor is statistically significant at the 5 % significance level. Table 6. The results of the Ordinal logistic Regression of the ques- tion “How often do you drink wine?”. Model R²N χ² df p 1 0.02 47.33 8 < .0001 Source: own processing. Table 7. Omnibus Likelihood Ratio Tests of the question “How often do you drink wine?”. Predictor χ² df p gender 15.05 1 0.0001 age 8.72 6 0.1897 education 11.09 1 0.0009 Source: own processing. Table 8. The results of the chi-square test of the question “Which wine do you prefer?”. Criterion χ2 Critical value Result gender 15.35246 5.991465 rejected age 29.48014 18.30704 rejected education 6.094705 5.991465 rejected income 20.56061 15.50731 rejected size of residence 14.64243 15.50731 not rejected Source: own processing. Table 9. The results of the multinomial Logistic Regression of the question “Which wine do you prefer?”. Model R²N χ² df p 2 0.05 62.57 24 < .0001 Source: own processing. 45Structure and development of the Czech wine market and foreign wine trade Nagelkerk’s pseudo R2 of 0.05 indicates a small effect on the explanatory variable. Table 10 summarizes the effect of each predictor on different categories of the dependent variable. It is clear from the table that at the standard 5 % significance lev- el, only the variables age and gender have a significant effect in the model. For the other variables we see statis- tically inconclusive effects. In connection with wine type selection, respond- ents were also asked about the preference for dry, sweet or semi-sweet wine. Most respondents indicated a pref- erence for semi-sweet wine (452 responses), the least respondents indicated dry wine (293 responses). Tested null hypothesis: - H05: The gender, age, education, income and size of resi- dence do not significantly predict the wine type preference (Dry, semi-sweet and sweet). The results of the chi-square test are shown in the following table (Table 11). The research showed the existence of a relationship between the preference of the wine type and the educa- tion and income of the respondents. Based on the results of the Chi-square test, a Mul- tinomial Logistic Regression model was again created. The reference category was the most numerous category – semi-sweet wine and only the variables education and income, which are statistically significant according to the Chi-square test, were included in the model. The results of the multinomial logistic regression are shown in the following Table 12. Based on the result of the Likelihood Ratio test, we accept the alternative hypothesis, i.e. at least one predic- tor is statistically significant at the 5 % significance lev- el. Nagelkerk’s pseudo R2 of 0.04 indicates a very small effect of the predictors on the explanatory variable. Table 13 summarizes the impact of each predictor on different categories of the dependent variable. The table shows that at the standard 5 % significance level, both variables have a significant effect in the model. The next research question tested was where people buy wine most often. In this question, respondents chose a supermarket, a wine shop, directly from a wine-maker or from fair-trade stores. As expected, the most frequent answer was a supermarket, followed by a wine shop. The lowest frequency of responses was for fair-trade stores. Subsequently, null hypothesis was established: - H06: The gender, age, education, income and size of resi- dence do not significantly predict the place of purchase of wine. The results of the chi-square test are shown in the following table (Table 14). The research has shown a relationship between the place of purchase of wine and age and the size of resi- dence. In her research, Chládková et al. [16] also showed the highest proportion of supermarkets as places of sale of wine (45.2% of respondents buy wine in supermarkets and 23.8% in wine shops). Based on the results of the Chi-square test, a Multi- nomial Logistic Regression model was again created. The Table 10. Omnibus Likelihood Ratio Tests of the question “Which wine do you prefer?” Predictor χ² df p age 23.15 12 0.0265 education 2.55 2 0.2801 gender 11.76 2 0.0028 Income 14.82 8 0.0628 Source: own processing. Table 11. The results of the chi-square test of the question “Which wine do you prefer?”. Criterion χ2 Critical value Result gender 0.808905 5.991465 not rejected age 6.010452 9.487729 not rejected education 20.22551 5.991465 rejected income 27.54793 15.50731 rejected size of residence 5.039119 15.50731 not rejected Source: own processing. Table 12. The results of the multinomial Logistic Regression of the question “Which wine do you prefer?” Model R²N χ² df p 3 0.03 41.67 10 < .0001 Source: own processing. Table 13. Omnibus Likelihood Ratio Tests of the question “Which wine do you prefer?” Predictor χ² df p education 12.57 2 0.0019 income 21.28 8 0.0064 Source: own processing. 46 Kamila Vesela, David Křížek, Lucie Severova reference category was the largest category – supermar- ket shopping and only the variables age and size of the municipality, which are statistically significant according to the Chi-square test, were included in the model. The results of the multinomial logistic regression are presented in the following Table 15. Based on the result of the Likelihood Ratio test, we accept the alternative hypothesis, i.e. at least one predictor is statistically significant at the 5 % significance level. How- ever, it should be noted that this significance level is very marginal. A Nagelkerk pseudo R2 of 0.02 indicates a very small effect of the predictors on the explanatory variable. Table 16 summarizes the impact of each predictor on different categories of the dependent variable. It is clear from the table that at the standard 5 % significance level, only municipality size has an effect on the depend- ent variable. The last question tested is the price range of pur- chased wines. Respondents chose from the following price categories: up to CZK 70, CZK 70–100, CZK 100– 150 and over CZK 150. From the point of view of abso- lute frequencies, people buy wine the most in the price category CZK 100–150, the least in the category up to CZK 70. Tested hypothesis: - H07: The gender, age, education, income and size of resi- dence do not significantly predict the choice of price catego- ry when buying wine. The results of the chi-square test are shown in the following table (Table 17). The research has shown the existence of a relation- ship between the price range of wine and gender, age, income and size of residence. Based on the Chi-square test results, an Ordi- nal Logistic Regression model was created. The refer- ence category was the last category – over CZK 150. All explanatory variables were included in the model except education, which is statistically insignificant according to the Chi-square test. The results of the ordinal logistic regression are shown in the following Table 18. Table 14. The results of the chi-square test of the question “Where do you buy wine most often?”. Criterion χ2 Critical value Result gender 0.990744 5.991465 not rejected age 21.45865 18.30704 rejected education 0.612048 5.991465 not rejected income 7.424061 9.487729 not rejected size of residence 16.73919 15.50731 rejected Source: own processing. Table 15. The results of the multinomial Logistic Regression of the question “Where do you buy wine most often?”. Model R²N χ² df p 4 0.02 31.79 20 0.0456 Source: own processing. Table 16. Omnibus Likelihood Ratio Tests of the question “Which wine do you prefer?” Predictor χ² df p size of residence 17.66 8 0.0239 age 13.00 12 0.3689 Source: own processing. Table 17. The results of the chi-square test of the question “In what price range do you most often buy wine?”. Criterion χ2 Critical value Result gender 17.41022 7.814728 rejected age 67.97642 12.59159 rejected education 3.231538 7.814728 not rejected income 36.23751 15.50731 rejected size of residence 137.8922 21.02607 rejected Source: own processing. Table 18. The results of the ordinal logistic regression of the ques- tion “In what price range do you most often buy wine?”. Model R²N χ² df p 5 0.02 36.69 15 0.0014 Source: own processing. Table 19. Omnibus Likelihood Ratio Tests of the question “In what price range do you most often buy wine?”. Predictor χ² df p Age 9.9 6 0.1288 Size of residence 3.63 4 0.4588 Gender 0.2 1 0.6566 Income 25.21 4 <.0001 Source: own processing. 47Structure and development of the Czech wine market and foreign wine trade Based on the result of the Likelihood Ratio test, we accept the alternative hypothesis, i.e. at least one predic- tor is statistically significant at the 5 % significance lev- el. Nagelkerk’s pseudo R2 of 0.02 indicates a very small effect of the predictors on the explanatory variable. Table 19 summarises the impact of the individual predictors. The table shows that at the standard 5% sig- nificance level, only income has an effect on the depend- ent variable. 4. DISCUSSION Wine is one of the most frequently consumed alco- holic beverages in the Czech Republic. There can be several motivations for wine consumption; Anchor and Lacinova [18] points out that one of the strong motives for wine consumption is social respect, which occurs especially in the young generation. However, it is pos- sible to go even further, and Oyinsey’s et al. [19] arti- cle, for example, is suggestive, as it takes into account a multidimensional experiential framework that seeks to uncover the dimensions that shape the experience of wine consumption. Over the last 20 years, the Czech Republic has shown significant and sustained growth in consumption, which is faster than growth in wine production. However, domestic production still exceeds wine consumption. The growing consumption of wine is also reflected in the growing demand for wine imports. However, this is not just a matter for the Czech Repub- lic, but as Bonn [20] shows, as awareness grows, so too does demand, globally [21]. The growing consumption of wine is also reflected in the growing demand for wine imports. From the point of view of the structure of the foreign wine trade, imports of sparkling wine are grow- ing the fastest, but the largest item of foreign trade is the import of still wine. Straková [22] adds that with the growing frequency of wine drinking, the consumption of Moravian and Czech wines prevails over foreign ones. Knowledge of the consumption and purchase motives of citizens of the Czech Republic, especially the young generation, is especially important for under- standing the Czech wine market, as these people are the current and future consumers of wine [23]. The research has shown the following conclusions: the frequency of wine drinking depends on gender, age and educa- tion. Here it can be further noted that gender not only affects consumption, but also preferences and customer satisfaction, for example Mitchell and Walsh [24], Atkin et al. [25]. The same is also true in the case of age- in addition to consumption, it has an impact on both of the above, for example Thach and Olsen [26] or Olsen et al. [27]. There is a relatively higher frequency of wine drinking among women than men, wine consumption increases with age and there is higher wine consump- tion among people with lower education. In the case of testing wine type preferences, dependence on gender, age, education and income was confirmed. Consump- tion is clearly dominated by the consumption of white wine. White wine is clearly preferred more by women, it also dominates across all age categories. Consumption of white wine also increases with increasing education and decreases with increasing age. On the fact that age can play an important role in preferences, for example, the article by Hammond et al. [28] which states age is important in wine consumption, but also in wine prefer- ences, on consumption behavior. Consumption of semi- sweet wines clearly dominates in the decision-making on wine consumption in relation to the sugar content in wine. The dependence of the consumption of individu- al types of wine on education and income was proven. With increasing education, the consumption of sweet wines decreases and consumer preferences are shifting in favour of dry wines. Consumption of dry wines also increases with income, whereas consumption of sweet wines decreases with increasing income. In terms of the place of purchase of wines, supermarkets clearly domi- nate. The research has shown a relationship between the choice of place to buy wines and age and the size of residence. With age, the amount of purchases from the wine-maker increases and the number of purchases in supermarkets decreases. Němcová and Stankova [29] confirms that the most common places of purchase for generation Y are supermarkets and wine shops. From the point of view of the price category of wines, the con- sumption of wines in the price range of CZK 100–150 dominates the most. The results confirm, among other things, the research of Chládková et al. [3], who identi- fied the most important factors influencing wine con- sumption in the form of consumer disposable income, product price and the existence of available substitutes (she considered beer in particular). Here, however, is the limit of this article, as it does not distinguish between the price a consumer would prefer when buying wine for themselves and when buying wine as a gift. As Cholette and Castaldi [30] point out, the price when buying wine as a gift tends to be higher. As Yu et al. [31] finds, this difference can be as much as ten times greater. If we compare the results of the presented research with the findings of the research carried out in Italy, we observe certain differences. Consumer preferences when buying wine in Italy using the “best-worst” scal- ing method have shown that direct, especially sensory experiences of consumers are key when choosing wine 48 Kamila Vesela, David Křížek, Lucie Severova in retail stores. The local statistical analysis showed that the age of consumers plays a role in the selection of wine in retail stores, the geographical location has not been proven [32]. Gil and Sánchez [33] analysed the factors shaping demand in Spain, especially between two differ- ent regions, Aragon and Navarre, using a weighted least squares approach. The research focused on three attrib- utes, namely the price, origin and vintage of the grapes. Surprisingly, the presented results (at least in compari- son with the results of research from the Czech Repub- lic) showed that the most important attribute of wine purchase is its origin, followed by the vintage. 5. CONCLUSIONS Although the Czech Republic is described as a beer country, the popularity of wine and its consump- tion shows clear and long-term stable growth. We also observed a positive development in the production of vines, which has a long tradition in the Czech Republic. The question of the future direction of the Czech wine market is promising and further development of the market in terms of production of domestic wine-mak- ers as well as overall consumption and growth of com- petition can be expected. Related to this is the need for wine producers to submit more to the wishes of demand (customer orientation) and thus the need to know the demand, its structure, preferences, expectations and fac- tors that affect it becomes more important. ACKNOWLEDGMENT This work was supported by the Faculty of Econom- ics and Management, Czech University of Life Sciences in Prague under Grant “The impact of climate change on the economic performance of the viticulture and wine-making sector in the Czech Republic” number 2020B0002.” REFERENCES [1] Bentzen, J., Smith, W. (2009): Wine Production in Denmark: Do the Characteristics of the Vineyards Affect the Chances for Awards? Available at http:// www.vdqs. net/search/chrono.asp?refer=recherche (accessed September 2020). [2] Marquardt, V., Hanf, J.H. (2012): Assessment of foreing influence on the Armenian wine business. Enometrica 5, 27–44. [3] Chládková, H., Gurská, S., Tomšík, P. (2009) The development of main factors of the wine demand. Agric. Econ. – Czech. 55(7), 321-326. [4] Kučerová, R., Žufan, P. (2008) Market position of selected competitors of the Czech wine market. Agric. Econ. – Czech. 54(7), 343-346. [5] Chládková, H. (2007): The competitive strength factors of the company. Agricultural Economics – Czech 53(7), 312–317. [6] Pyšný, T., Pošvář, Z., Gurská, S. (2007): Analysis of selected demand factors of wine market of the Czech Republic. Agricultural Economics – Czech 53(7), 304–311. [7] Butkus, M., Cibulskiene, D., Maciulyte-Sniukiene, A., & Matuzeviciute, K. (2018). What is the evo- lution of convergence in the EU? Decomposing EU disparities up to NUTS 3 level. Sustainability, 10(5), 1552. [8] Škorpíková, A. (2003): Podnikání na trhu vína v ČR – případová studie. PEF MZLU, Brno. http// www.eu.cz [9] Melecký, L., Staničková, M. (2011): Hodnoceni konkurenceschopnosti regionů Česke republiky v  kontextu Lisabonske strategie. (Evaluation of the Czech Republic regions competitivenes in the context of the Lisbon strategy.) Ekonomicka revue – Central European Review of Economic Issues 14, 183–200. [10] Chládková, H., Gurská, S., Tomšík, P. (2011): Wine Environment in the Czech Republic. Avail- able at http://w w w.vdqs.net/search/chrono. asp?refer=recherche (accessed September 2020). [11] Verner, T. (2011): Ekonomicky růst, konkurenc- eschopnost a kvalita života. (Economic growth, competitiveness and quality of life.) Ekonomicka revue – Central European Review of Economic Issues 14, 157–166. [12] Svatoš, M., Smutka, L. (2012): Development of agricultural trade and competitiveness of the com- modity structures of individual countries of the Visegrad Group. Agricultural Economics – Czech 58, 222–238. [13] Tomšík, P., Prokeš, M. (2011) New strategic alli- ances of wine producers in the Czech Republic. AGRIC. ECON. – CZECH. 57(12), 573–579. [14] Tomšík, P., Sedlo, J. (2013) Management of wine production with regard to its implementation into the wine markets of the Czech Republic. AGRIC. ECON. – CZECH. 59(5), 202-210. [15] Hambalková, M. (2006) The factors of competitive- ness and the quantification of their impact on the export efficiency of grape and wine in the Slovak Republic. Agric. Econ. – Czech. 52(8), 389-394. 49Structure and development of the Czech wine market and foreign wine trade [16] Chládková, H., Pošvář, Z., Žufan, P. (2004) Con- sumer habits in the Czech wine market. Agric. Econ. – Czech. 50(7), 323-330. [17] Kelley, K., Hyde, J. and Bruwer, J. (2015) U.S. wine consumer preferences for bottle characteristics, back label extrinsic cues and wine composition: A conjoint analysis, Asia Pacific Journal of Mar- keting and Logistics 27(4), 516-534. https://doi. org/10.1108/APJML-09-2014-0140 [18] Anchor, R. J., Lacinová, T. (2015) Czech wine con- sumers: maturing with age. Ekonomie a Manage- ment. 18(1), 169-182. DOI: 10.15240/tul/001/2015- 1-013. [19] Oyinseye, P., Suárez, A., Saldaña, E., Fernández- Zurbano, P., Valentin, D., & Sáenz-Navajas, M. P. (2022). Multidimensional representation of wine drinking experience: Effects of the level of consum- ers’ expertise and involvement.  Food Quality and Preference,  98, https://doi.org/104536.  10.1016/j. foodqual.2022.104536 [20] Bonn, M. A., Cho, M., & Um, H. (2018). The evo- lution of wine research: A 26 year historical exami- nation of topics, trends and future direction.  Inter- national Journal of Contemporary Hospitality Management. [21] Bruwer, J., & McCutcheon, E. (2017). Marketing implications from a behaviourism perspective of consumption dynamics and socio-demographics of wine consumers.  Asia Pacific Journal of Marketing and Logistics. [22] Straková, A. (2017). Výroční zpráva Vinařského fondu za rok 2016. Brno: Vinařský fond. Retrieved from http://vinarskyfond.cz/wpcontent/ uploads/2017/05/VZ-Vinarsky-fond-2016-web.pdf. [23] VinIntell. (2013). Future Scenarios for the South African Wine Industry: Part 5 – Generation Y: the new market colossus. SAWIS. Retrieved from http://www.sawis.co.za/info/download/VinIntell_ June_2013_part_5_issue_16.pdf. [24] Mitchell, V. W., & Walsh, G. (2004). Gender differ- ences in German consumer decision‐making styles. Journal of Consumer Behaviour: An International Research Review, 3(4), 331-346. [25] Atkin, T., Nowak, L., & Garcia, R. (2007). Women wine consumers: information search and retailing implications. International Journal of Wine Busi- ness Research. [26] Thach, E. C., & Olsen, J. E. (2006). Market segment analysis to target young adult wine drinkers. Agri- business: An International Journal, 22(3), 307-322. [27] Olsen, J. E., Thach And, L., & Nowak, L. (2007). Wine for my generation: exploring how US wine consumers are socialized to wine. Journal of Wine Research, 18(1), 1-18. [28] Hammond, R., Velikova, N., & Dodd, T. H. (2013). Information sources used by millennial restaurant wine consumers.  Journal of Foodservice Business Research, 16(5), 468-485. [29] Němcová, J., Staňková, P. (2019). Factors influ- encing consumer behaviour of generation Y on the Czech wine market. Ekonomie a Management 22(4), 145-161. DOI: 10.15240/tul/001/2019-4-010. [31] Cholette, S., & Castaldi, R. (2005). Analyzing the US retail wine market using price and consumer segmentation models (referred). San Francisco University (24). Recuperado de http://academy- ofwinebusiness.com/wpcontent/uploads/2010/05/ AnalyzingUSRetail. pdf. [32] Yu, Y., Sun, H., Goodman, S., Chen, S., & Ma, H. (2009). Chinese choices: a survey of wine consum- ers in Beijing.  International Journal of Wine Busi- ness Research. [32] Gil, J.M. and Sánchez, M. (1997), “Consumer pref- erences for wine attributes: a conjoint approach”, British Food Journal 99(1), 3-11. https://doi. org/10.1108/00070709710158825 Wine Economics and Policy Volume 12, Issue 1 – 2023 Firenze University Press State of the International Wine Market in 2022: New market trends for wines require new strategies Rafael Del Rey1, Simone Loose2,* Does anyone read my papers? The gap between academic consumer research and the real (wine) world Riccardo Vecchio A certification for natural wine? A comparative analysis of consumer drivers in Italy and Spain Eva Parga Dans1,*, Riccardo Vecchio2, Azzurra Annunziata3, Pablo Alonso González4, Raimundo Otero Enríquez5 Structure and development of the Czech wine market and foreign wine trade Kamila Vesela, David Křížek*, Lucie Severova The impact of alternative packaging on the life cycle of wine on tap Stefano Massaglia1, Tibor Verduna1, Vincenzo Varchetta2, Filippo Brun1, Simone Blanc1,* The new CAP and the challenge of sustainability: a synthetic indicator for the Italian wine sector Roberta Sardone1,*, Simonetta De Leo1, Davide Longhitano2, Roberto Henke1 Analysis of the 2007-2008 Hérault premiumized grubbing-up campaign: a tool to better understand Fischer-Boel’s 2008-2011 grubbing-up campaigns and the desire in 2022 to reintroduce locally premiumized grub-ups Étienne Montaigne1, Samson Zadmehran2,*, Alfredo Coelho3, Yacine Messaoudène4 The impact of fees on customer purchasing behavior and beliefs in winery tasting rooms: A scoping review John C. Spence