PRIVATE CONSUMPTION IN THE EUROPEAN UNION: A COMPARATIVE STUDY

The aim of this paper is to study consumption of households from an economic and cultural perspective in the European Union with 28 Member States during the period 2010-2019. For this purpose, we compared the Eastern European countries, dominated by rapid economic growth and development with the Western European countries, which represent the most developed countries in the EU-28. From this perspective, we proposed a multidimensional analysis of consumption that includes macroeconomic indicators of households’ wealth, which strongly influence their consumption together with an overview on expenditure by consumption purpose. Moreover, we have also considered Hofstede’s cultural dimension theory based initially on four cultural dimensions (power distance, individualism versus collectivism, masculinity versus femininity, and uncertainty avoidance) to observe the impact national culture plays on households’ consumption in Eastern and Western European countries tracking the historical changes of these countries. Our methodological approach consisted in descriptive and inferential statistics based on the selected economic and cultural indicators. Pearson’s product-moment correlations were calculated to assess the correlations between the variables. Our analysis shows that the level of wealth is lower in Eastern European countries compared to Western Europe, which influences significantly the private consumption in these countries. Moreover, the systematic differences of national culture between Eastern and Western Europe influence strongly the private consumption of their population. Results of this paper indicate that in Eastern European countries the highest share of expenditure is allocated to primary needs such as food, non-alcoholic beverages, alcoholic beverages and cigarettes to the detriment of health, education, recreation and culture.


INTRODUCTION
Our society is called a consumerist society as the level of consumption of goods and services nowadays has never been seen in the history of humanity (Matsuyama 2002:5-7). The gradual transition towards consumerism started after and it has triggered progress in all areas of activity, which lead to a lifestyle of individuals inclined towards wealth.
Private consumption represents the household's consumption of goods and services. The households' consumption accounts for around 50% of EU-28 Gross Domestic Product (GDP) compared to around 20% for government consumption (Eurostat, 2019). The assignation of financial resources towards private consumption implies a reduction of the personal savings, and, in some cases even indebtedness of individuals and households. Households' income determines the level of their consumption and the category of goods and services towards which they decide allocate their budget.
The private consumption is very limited in those households in which the income does not cover the expenses of their basic needs. According to Engel's law, the wealthier households allocate a smaller share of their budget for food, which is a basic need, assigning their resources to more sophisticated needs.
Households' consumption in the European Union is still divergent triggering numerous differences between countries. Factors that drive households' consumption are triggered by economic features such as disposable income and individual consumption, and socio-cultural characteristics such as habits and cultures. In this paper, we have considered Hofstede's cultural dimension theory based initially on four cultural dimensions (power distance, individualism versus collectivism, masculinity versus femininity and uncertainty avoidance) to observe the impact national culture has on households' consumption in EU-28 Member States during 2010-2019. For this purpose, we have drawn statistical comparisons between the Western European countries, the most developed countries in the EU-28 and Eastern European countries, developing countries, which joined the EU after 2004.
This paper is structured in five parts. The first part contains the introduction, which indicates the research question. The second part of the paper covers the literature review of economic and cultural influences of households' consumption in the European Union. The third part of the study describes the research methodology including the data sources. The fourth part of the paper presents the results and discussions derived from the study. Finally, the fifth part shows the conclusions of the study.

LITERATURE REVIEW
Some authors define consumption as "spending for survival or enjoyment as opposite to providing for future production" (Black 1997:84). Recent research (Sadik-Zada & Loewenstein 2018:196-204) reveals the direct relationship between disposable income and consumption. Other authors argue that income effects show differences in the spending patterns of affluent households rise (Chai et al. 2015:423-440) and that changes in household wealth impact on consumption (Jawadi et al. 2017:849). Furthermore, consumption is shaped by the interaction between the national context and individual lifestyle preferences (Thøgersen 2017:16-25). Moreover, households' consumption is a complex process, which represents the result of social, cultural and economic influences (Zukin & Maguire 2004:174). Other authors (Schor 2002:3-4) indicate that motivations of consumption are social, these deriving from the need to keep pace with the level of consumption promoted by the society rather than with objective individual needs. Some authors (Firat et al. 2013:199) conclude that culture is one of the most important factors affecting attitudes, behaviors and lifestyles of households' consumption.
If we consider the socio-economic context of the countries in the European Union, we can distinguish two different societies: Western Europe and Eastern Europe. On one hand, the socioeconomic evolution of Eastern European countries produced by the collapse of the communist regime triggered a significant change in the political regime. This implied a major transition from the planned economy towards a market economy (Lipton et al. 1990:6-79). On the other hand, countries from the Western Europe are more economically developed having an individualistic culture, whereas Eastern European countries are developing at a rapid pace being more traditional (Trentmann 2004:378). Some authors indicate the differences in the structure of households' consumption appeared due to different levels of socio-economic development in different countries along with historical, geographical and cultural factors (Kozera et al. 2013:293-303). For some categories of consumption such as clothing, footwear, miscellaneous goods and services Eastern European countries before joining the EU spent less than Western European countries (Arnotte 1999

RESEARCH METHODOLOGY
The aim of this paper is to analyze the households' consumption in the European Union and to identify some economic and social variables that influence consumption. In our analysis we refer to the period 2010-2019 based on the data availability of the indicators analyzed. In this paper, we have used descriptive statistics to correlate economic indicators such as the Adjusted gross disposable income of households per capita, the Actual Individual Consumption and Households' expenditure by consumption purpose with the cultural factors based on Hofstede's initial model of national culture. Both economic indicators used in this paper are expressed in Purchasing Power Standards (PPS) to include the price level differences across the countries subject to our analysis.
Adjusted gross disposable income of households per capita reflects "the purchasing power of households and their ability to invest in goods and services or save for the future" (Eurostat, 2019). This indicator includes the flows of services households receive free of charge from governments (e.g. education, health, etc.). In this paper, we used this indicator as it reflects the disposable income of households available for consumption and/or savings.
Actual Individual Consumption (AIC) measures all the goods and services consumed by households including the services provided by government or non-profit organizations (e.g. education, health, etc.). This indicator is preferred in comparative studies as indicator of households' material welfare. In this paper, we focus only on households' consumption, therefore we have excluded from our analysis the services provided by government or non-profit organizations.
Households' expenditure by consumption purpose depicts the allocation of income aggregated at the national level to the main categories of spending such as food & non-alcoholic beverages, clothing & footwear, housing, health, education, recreation and culture.
From the side of cultural factors that influence consumption, we relied on Hofstede's model of national culture (Hofstede 1980:15-41) that initially comprised four dimensions: power distance, individualism versus collectivism, masculinity versus femininity and uncertainty avoidance.
Power distance index (PDI) reflects the way in which a society reacts to inequalities among its members. A high power distance index is characteristic for societies in which people accept differences and hierarchical order, while in countries with a low power distance people with less power strive to minimize the inequalities in the distribution of power.
Individualism versus collectivism (IDV) shows whether the members of a society are inclined to act only for their own welfare and for their families (high index values) or on the contrary it is a society in which members of different groups help each other selflessly.
Masculinity versus femininity (MAS) differentiates between societies oriented towards achievement, heroism, material rewards for success (high values correspond to masculinity) and those in which cooperation and modesty are appreciated values (low values correspond to femininity).
Uncertainty avoidance index (UAI) reflects how uncomfortable are the members of a society with ambiguity and uncertainty. Societies with a high UAI are more intolerant and respect rigid codes of behavior, while those with a low UAI have a more relaxed attitude and accept more easily uncertain situations.
In our analysis we verified the following hypotheses: H1: Actual Individual Consumption is positively correlated with the Adjusted gross disposable income of households per capita.
H2: Households' expenditure by consumption purpose is positively correlated with the Adjusted gross disposable income of households per capita. .
H3: Actual Individual Consumption is positively correlated with the Hofstede's 4 dimensions of national culture.
H4: Households' expenditure by consumption purpose is correlated with the Hofstede's 4 dimensions of national culture.

June No. 1/2021
The statistical approach applied to verify these hypotheses is the Pearson's product-moment correlation analysis. Moreover, the results and discussions derived from our analysis are presented in section 4 of this paper.

RESULTS AND DISCUSSIONS
The total gross disposable income of households in EU-28 amounted to EUR 9 781 billion in 2018 (Eurostat, 2019). Figure 1 reflects the evolution of the Adjusted gross household disposable income per capita in EU-28 during the period 2010-2019 (data for Malta and UK was not available). In this paper, we used the data converted in purchasing power standards (PPS), which allowed us to draw an accurate comparison between all the EU-28 Member States. The mean Adjusted gross household disposable income per capita in EU-28 had an upward movement during the whole period analyzed, increasing from 19 650 PPS (2010) to 23 599 PPS (2019). However, major differences can be observed between the Member States from Eastern and Western Europe. The major difference is between Luxembourg and Bulgaria, the countries with the highest and respectively lowest value in terms of Adjusted gross income per capita. In 2010, Luxembourg exceeded almost 4 times the Adjusted gross disposable income per capita compared to Bulgaria. In 2019, the country that has exceeded the EU-28 average was Germany (+28,5 %), followed by Austria (+19,4 %), Belgium (+14,8 %), Netherlands (+12,3 %), France (+10,8 %), Finland, Sweden and Denmark (less than 10 %) (data for Luxembourg was not available for 2019, but in 2018 it registered 41,2 % more than the EU-28 average). At the opposite pole, we find the Eastern European countries as eight countries recorded values of 20 % or more below the EU-28 average (data for Bulgaria was not available, but in 2017, the most recent available year, it registered 46,1 % of the EU-  The evolution of the Actual Individual Consumption (AIC) is analyzed in this paper as an indicator for the material welfare of households at the European Union level for the period 2010-2019 ( Figure 2). As observed in Figure 1, in Figure 2 we remark a similar pattern in households' welfare with significant differences between the Eastern and Western European countries. The mean AIC for EU-28 increased from 17 600 PPS (2010) to 21 000 PPS (2019). The countries that exceed the EU-28 AIC average are Luxembourg (+31,9 %), Germany (+20 %), Austria (+15,2 %), Denmark (+13,8 %), Netherlands (+12,4 %), United Kingdom (+11, 4%) followed by Finland (+11 %) and France (+7,1 %). The countries with the lowest AIC in EU are Bulgaria (-42,9 % from the EU-28 mean), Croatia (-35,7 %), Hungary (-33,8 %), Slovakia (-31,9 %), Latvia (-35,5 %) followed by Estonia, Greece, Poland and Romania with less than 30 % below the EU-28 AIC average. Slovenia, Czech Republic, Malta, and Portugal registered values of AIC with 20 % below the EU-28 average, while Spain, Ireland, Italy, Cyprus and Lithuania had values less than 10 % below the EU-28 mean.
The analysis of Figures 1 and 2 indicate that Eastern European countries position themselves lower on both economic indicators used in this study compared to the Western European countries. The differences in the welfare of households between the EU-28 Member States reveal challenges that developing countries from Eastern Europe strive to attain in order to reach the development level of Western European countries.   In this paper, we considered the hypothesis following which the AIC and the distribution of households' expenditure across different categories of consumption is not only influenced by the economic factors, but also by national cultural dimensions. For this purpose, we correlated the Actual Individual Consumption and Households' expenditure by consumption purpose with the Adjusted gross disposable income, and with the 4 dimensions of national culture proposed initially by Hofstede. The results of the correlation tests (Table 2) show that AIC is positively correlated with Adjusted gross disposable income of households per capita (p-value<0.0001) and IDV (p-value<0.01), and negatively correlated with PDI (p-value<0.001) and UAI (p-value=0.05). The correlation between AIC and MAS is not statistically significant. Thus, countries with higher income per capita and with individualistic population register higher values on AIC. On the contrary, countries in which people accept inequalities between their members and avoid uncertainty, have lower values on AIC.
Regarding the correlations with households' expenditure by consumption purposes we observed that the Adjusted gross disposable income of households per capita is positively correlated with Furnishings, household equipment and routine household maintenance (p-value<0.01), Miscellaneous goods and services (p-value<0.0001), and, negatively correlated with Food and non-alcoholic beverages (p-value<0.0001), Alcoholic beverages, tobacco and narcotics (p-value<0.05), Communications (p-value<0.0001), and Education (p-value<0.05). PDI is positively correlated with Food and non-alcoholic beverages (p-value<0.01), Communications (p-value<0.01), and negatively correlated with Recreation and culture (p-value<0.05). These findings suggest that in societies in which people accept inequalities between their members, less money will be spent on recreational and cultural activities and more money on Food and non-alcoholic beverages and Communications. If we consider that people with less power have also lower levels of income, then these results are relevant and intuitive. Individualism versus collectivism is positively correlated with Housing, water, electricity, gas and other fuels (p-value<0.05), Recreation and culture (p-value<0.05), Miscellaneous goods and services (p-value<0.05) and negatively correlated with Food and non-alcoholic beverages (p-value=0.01) and Communications (p-value<0.01). Masculinity versus femininity index is not significantly correlated with any of the variables included in the analysis. In societies in which people have a high level of uncertainty avoidance, more is spent on Food and non-alcoholic beverages (p<0.05), Health (p<0.05), and less on Housing, water, electricity, gas and other fuels (p-value=0.01) and Recreation and culture (p-value<0.001).

CONCLUSIONS
In this paper, we analyzed the households' consumption in the European Union with 28 Member States during the period 2010-2019 from an economic and cultural perspective. In the first part of our analysis, we described the evolution of two macroeconomic indicators, the Adjusted gross disposable income of households per capita and the Actual Individual Consumption per capita. The results show that the population from the Eastern Europe is less wealthy than the population from the Western countries, since they recorded lower values on both economic indicators. The countries with the lowest Adjusted gross disposable income per capita are Bulgaria, Estonia, Poland, Slovakia, Romania and Hungary, while the countries with the highest values on this indicator are Luxembourg, Germany, Austria, Belgium and Netherlands. The EU-28 average of the Actual Individual Consumption increased significantly during the period analyzed from 17600 PPS per capita in 2010 to 21 000 PPS per capita in 2019. The countries with the lowest AIC are the Eastern European countries, namely Bulgaria, Croatia, Hungary, Slovakia and Latvia. At the opposite pole, there are the Western European countries, such as Luxembourg, Germany, Austria, Denmark and Netherlands. Regarding the household expenditure by consumption purpose, we observed that Eastern European countries households allocate higher shares of their total households' expenditure to the basic categories of consumption such as food, alcoholic beverages, tobacco in the detriment of education, recreational and cultural activities. The population living in the Western Europe allocate higher shares of their budget than the population living in the Eastern Europe for categories of consumption such as Housing, water, electricity, gas & other fuels, Clothing and footwear, Recreation and culture, Restaurants and hotels. We hypothesized that the differences in consumption are also related with the cultural differences between the two groups of countries, Western Europe and Eastern Europe.
In the second part of our analysis, we verified four hypotheses regarding the correlation between the Actual Individual Consumption and Households' expenditure by consumption purpose on one side and the Adjusted gross disposable income and Hofstede's 4 cultural dimensions on the other side. We used the cultural dimensions in our analysis since they also influence consumption besides the economic factors. According to the correlation tests, the first hypothesis is accepted, since we obtained a positive correlation between the Adjusted gross disposable income and the Actual Individual Consumption. The second hypothesis is only partially accepted because only some categories of consumption expenditures are correlated with the Adjusted gross disposable income. The third hypothesis is also partially accepted: 3 dimensions of 4 are correlated with AIC (Power distance index, Individualism versus collectivism, and Uncertainty avoidance index. The fourth hypothesis is only partially true, as in the previous case, Hofstede's dimensions are correlated only with some categories of consumption expenditures.
Our study creates the grounds for future research of this topic, since the relation between culture & consumption is complex in nature and it deserves to be substantiated empirically. Therefore, for future research, we propose to apply more advanced methodology for statistical analysis, such as the structural equation modelling that will allow assessing not only the correlation but also the causation between various variables.