DIFFERENCES AND SIMILARITIES IN PATTERNS OF AGEING SOCIETY IN THE EUROPEAN UNION

45 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics DIFFERENCES AND SIMILARITIES IN PATTERNS OF AGEING SOCIETY IN THE EUROPEAN UNION Denisa Kočanová1, Viliam Kováč2, Vitaliy Serzhanov3, Ján Buleca4 1 Technical University of Košice, Faculty of Economics, Department of Finance, Slovakia, ORCID: 000-0001-6109-6657, [email protected]; 2 Technical University of Košice, Faculty of Mining, Ecology, Process Control and Geotechnologies, Slovakia, ORCID: 0000-0002-5265-9005, [email protected]; 3 Uzhhorod National University, Faculty of Economics, Department of Finance and Banking, Ukraine, ORCID: 0000-0002-0577-4422, vitaliy[email protected]; 4 Technical University of Košice, Faculty of Economics, Department of Finance, Slovakia, ORCID: 0000-0002-6613-2167, [email protected]. Abstract: Population ageing is a demographic issue that emphasises the need to be interested in the lives of the most vulnerable population group: the elderly population. The paper investigates the ageing process and their relations among the European Union member countries from 2009 to 2019. These countries are assessed and dispersed to the appropriate clusters according to several indicators related to the areas that affect the lives of the elderly population: namely, the health status, the labour market conditions, and financial security. The focus is on the age group 55 years and over as it is a disadvantaged age group in the job application process regarding ageing society. It is a significant aspect of public finance system. The European Union Statistics on Income and Living Conditions, the Labour Force Survey, and the European System of Integrated Social Protection Statistics data are involved. The quantitative approaches are applied in the cluster analysis and followed by the panel data linear regression analysis. The dendrograms visualise the three clusters representing the mutual relations and the ageing patterns among the explored countries. The heat maps are created to prove the potential relations among the observed countries. The panel regression model demonstrates that the three variables – part-time employment, the income inequality, and the material and social deprivation – are statistically significant in all the regression models for the whole area and the three clusters. The analytical outcome could be applied as a valuable resource to government and national representatives. It can help identify the objectionable determinants for a custom policy and implement appropriate measures to improve the situation of the elderly population. Keywords: Ageing, life expectancy, cluster analysis, regression analysis, European Union. JEL Classification: C33, I14, J14. APA Style Citation: Kočanová, D., Kováč, V., Serzhanov, V., & Buleca, J. (2023). Differences and Similarities in Patterns of Ageing Society in the European Union. E&M Economics and Management, 26(1), 45–64. https://doi.org/10.15240/tul/001/2023-1-003 Introduction Population ageing currently represents a phenomenon that is occurring around the world. It can also be defined as a consequence of the fertility rate decline and the increasing life ex pec tancy, resulting in an increasing number and a proportion of the population in the post-productive age. For the first time, the E+M_01_2023.indb 45 28.2.2023 10:02:35 46 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics elderly population will be predominant over the younger people (World Health Organization, 2020). The on going demographic changes are going to characterise the upcoming decades. This process will affect the different areas from the population’s health status through the health systems, the conditions in the labour market, the changing consumption patterns, and the need to provide the system reforms related to higher demand for public resources and finance. Ageing creates deep pressure on fiscal sustainability. It will put unprecedented stress on public finance to fund the pension system, the health system and the long-term care expenditures (Organisation for Economic Co-operation and Development, 2019). The burden on public finance lies precisely in need to increase government spending on the pension system, which plays a crucial role in ensuring the living standard requirements of the elderly population. Unfortunately, one in seven pensioners is at risk of poverty in the European Union member countries nowadays (Eurostat, 2021a). Ageing is mainly associated with a growing number of the population in the post-productive age. Therefore, it is essential to know the elementary characteristics and the socioeconomic indicators of this age group. An overview of the recent studies shows that the health status, labour conditions and financial security appear to be the most critical areas for the elderly population. This paper contributes to the current scientific literature by evaluating the ageing process and the demographic changes in the European Union territory. It classifies its member countries with similar ageing characteristics into different clusters. It also outlines the financial, health, and social indicators with the highest importance for the lives of the elderly population. All these three aspects are important as the public resources spent in the fields which these aspects are related to, that is, mainly public and government spending. The main objective of this paper is to detect a potential mutual relation between the European Union member countries as Europe is characterised by almost the oldest population in the world. If this were proved, the purpose would be to review which of the explored countries behave similarly and hence; the analogous policy could be applied for them. Above and beyond this, the development of the last ten years is examined too. Therefore, a partial goal is to analyse the selected variables related to the ageing population and cluster the observed countries according to this. Such a process can reveal the possible common succession of the steps in order to prepare the social conditions for all the age groups of the population in the given countries. The main research question is to reveal the relations related to the explored dimensions. Fundamentally, the hypotheses are stated in a way that the involved variables behave statistically significant for the explanatory variable representing old-age dependency. This represents an aspect of the public finance system that plays a key role in the financing processes. It creates a part of the whole issue of ageing society as this phenomenon covers many fields of life. Moreover, this should create a basement for a potential mutual public policy of the institutions of the European Union. Hence, the rules accepted by the European Union are not only funded by higher amount of financial resources, but also are backed by stronger legislative. In the submitted paper, an investigation of the impact of given variables is carried out with its significance in the particular countries. The paper is structured as follows: the second section offers the literature review, the third section describes the data, and the fourth section the methodology applied. The analytical outcome of the discussion is demonstrated in the fifth section. Finally, the sixth section represents the conclusion with a summary of the obtained findings. 1. Theoretical Background The demographic changes are very rapid globally, and the sheer number of pensioners is increasing faster than it is often thought (Officer et al., 2020). The developed world is subject to two simultaneous processes. Population growth is slowing, and the population’s age structure is changing in many countries (Broniatowska, 2019). By 2030, a quarter of the developed world’s population will be of age over 65 years (United Nations, 2019b). Moreover, the European continent possesses one of the oldest populations, as many European countries have the lowest fertility rates and the highest life expectancy rates globally (Marois et al., 2020). Numerous previous research was centred on the demography changes and process of population ageing in the countries of the European Union (Cristea et al., 2020; Ebbinghaus, E+M_01_2023.indb 46 28.2.2023 10:02:35 47 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics 2021; Estevens, 2018; Gómez-Costilla et al., 2021; Jakovljevic et al., 2018; Kluge et al., 2019; Lutz et al., 2019; Pascual-Saet et al., 2020; Sanderson & Scherbov, 2016; Skibiński, 2017). Most studies confirm that demographic changes are a challenge for many policy areas: labour market, health and long-term care, lifelong learning, family policy, social protection, and pension systems. A population decline can be examined through several aspects to explore its potential consequences (Coleman & Rowthorn, 2011; Fukuda & Okumura, 2020; Janus et al., 2022; Kikuchi et al., 2022). The fertility rate level itself cannot illustrate the whole situation because several other dimensions should be considered (Striessnig & Lutz, 2013). This field needs new approaches to quantify the crucial figures to determine the future policy (Sanderson & Scherbov, 2015). The scientific literature also emphasises the importance of studying the life of the elderly and focuses on healthy and active ageing. On the contrary, attention is also paid to the problems of social security and increased demand for pension payments. These are the result of the state confirmed by the Organisation for Economic Co-operation and Development (2019) that old-age dependency ratios will rise in all the 27 European Union member countries in the following decades, putting the financing of adequate pensions, health, and long-term care under high pressure. The increased burden on the social security system represents the expected consequence of an ageing population from the point of view of economics (Skibiński, 2017). The critical challenges of ageing regarding the way it shapes society are the financial security in retirement, the disease burden in old age, and the shrinking labour force in the productive age (Aguilar-Palacio et al., 2018; Kratt & Kirnos, 2020; Park et al., 2022). These challenges are domains included in the Active Ageing Index (Ortega, 2020; United Nations, 2019a). Lyons et al. (2018) focused on the impact of population ageing on the financial security of households that are most like to be vulnerable. Financial inclusion and technological usage are considered essential tools to ensure the financial security of the elderly. Also, financial security tends to be linked to the population ageing crisis (Khan, 2019; Mura et al., 2019; Pascual-Saez et al., 2020; Rupeika-Apoga & Thalassinos, 2019). In the context of the labour market situation, there are questions of who, how many and with what skills will work in Europe in the upcoming decades (Lutz et al., 2019). The ageing process and retirement significantly transform the situation in the labour market. To encourage the elderly to work longer, postpone retirement and provide them with the opportunity to be employed part-time seems to be the apparent policy solution (Moen, 2020; Sewdas et al., 2017). However, internal policy solutions such as increased retirement age or longer working hours were unsuccessful in many countries. Comparing differences of the role of retirement by the motive or the form for retirement claimed that the fully retired population showed a significantly lower sense of purpose than those who are still working or are just partially retired (Best & Hill, 2020). In recent years, several studies have been focused on the factors influencing the quality of life and life satisfaction of the elderlies (Leeuwen et al., 2019). For evaluating the life satisfaction of elderlies, it is crucial to recognise the factors that influence their quality of life, emphasising that older people tend to prefer the quality of life over a long period. Several studies found a negative association of age and health of elderlies, but a positive association with the difficulties and the limitations in daily activities. Although, this is disputable because of the selection bias occurrence. Some have also proved a direct impact of social support on self-perceived health, especially among the elderly (Giang et al., 2020; Le et al., 2020; Loichinger & Pothisiri, 2018). Also, the depressive symptoms are markedly connected to the quality of life (Bornet et al., 2017). Life satisfaction of the elderly population is negatively affected even in the case of low financial or material support, when they suffer from emotional stress or when they lose support provided by children (Liu et al., 2019). Hence, it is recommended to give special attention to studying the social differences in life expectancy at the age of 50 or 65 years and disability-free life expectancy. The aim is to estimate the contributions from the disability and mortality effects through the differences between the income groups. The gradient of income inequality is sharper for healthy life years than life expectancy. As E+M_01_2023.indb 47 28.2.2023 10:02:35 48 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics income inequality is increasing, disability-free life expectancy by income appears to be an important indicator that can monitor social inequality in the accretive share of elderlies (Brønnum-Hansen et al., 2021). Ageing also results in increased expenditures on healthcare services and welfare (Li et al., 2020). Therefore, it is valuable to focus on the lives of the elderly population, identify the most likely influential factors, and find out the purpose of retiring as it is connected to many physical and mental health implications (Best & Hill, 2020). 2. Methodology Several quantitative methods are applied, focusing on the cluster analysis and the sensitivity analysis in the form of panel data regression analysis as the primary approaches. Firstly, to provide that data are mutually comparable, the normalisation of the data is necessary. The maxima are the model outputs, while the minima are defined as the lowest performance benchmarks, and thus, they are determined as the worst possible scenario. The indicators are standardised into a consistent scale from zero to one according to Williamson and Piattoeva (2018) and Grannis et al. (2019) and proximity-to-target methodology: (1) where: Ixi – the standardised value of the given indicator; xi – i-th value of the variable to be standardised. If the indicator growth represents an undesirable trend, the following modification is ready to be applied: (2) where: the comprised variables possess the same meaning as the previous equation. Secondly, the similarity of the territories is computed through the Euclidean distance: (3) where: c1 – the first country; c2 – the second country; D(c1, c2) – the mutual Euclidean distance of the c1 country and the c2 country; c1x – the x coordinate of the c1 country; c2x – the x coordinate of the c2 country; c1y – the y coordinate of the c1 country; c2y – the y coordinate of the c2 country. Thirdly, the clusters are determined according to the following methods:  The Ball-Hall index (Ball & Hall, 1965);  The McClain-Rao index (McClain & Rao, 1975);  The point-biserial correlation coefficient (Milligan, 1981). The Ball-Hall index is computed in this way: (4) where: BHI – the Ball-Hall index; n – a number of the countries; i – the particular country; vi – a vector of the i-th country in the particular cluster; ci – a centroid of the i-th cluster; t – a modification parameter. The McClain-Rao index is calculated as follows: (5) where: MRI – the McClain-Rao index; Dw – a sum of the within cluster distances; Pw – a number of the country pairs of the observations belonging to the within cluster; Db – a sum of the between cluster distances; Pb – a number of the country pairs of the observations belonging to the between cluster. The point-biserial correlation coefficient is computed like this: (6) where: PBCC – the point-biserial correlation coefficient; Dw – a sum of the within cluster distances; Db – a sum of the between cluster distances; Pw – a number of the country pairs of the observations belonging to the within cluster; Pb – a number of the country pairs of the observations belonging to the between cluster; SDD – a standard deviation of all the distances. The construction of the clusters is based on Ward’s minimum variance method. The dendrograms and heat maps represent the graphical outcome of the cluster analysis. E+M_01_2023.indb 48 28.2.2023 10:02:35 49 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics The second methodological approach applied is sensitivity analysis. Within it, the regression model of a linear form for the panel data is constructed. The input data set for the modelling process possesses a form of panel data, including the two dimensions: a territorial one and a time one. The first difference method is applied in the linear regression. The regression analysis is carried out with a presence of a constant value. The sequential elimination method is the primary modelling technique, so the variable with the lowest statistical significance is excluded from the other modelling process. The elimination factor is represented by the p-value of the particular independent variable. The sequential elimination is related to the elementary altogether model for the whole set of the explored countries. It implies that the cluster regression models aimed at the clusters are adapted to the elementary model. Therefore, it involves the variables in the final model of the modelling row that was not compulsory to meet the statistical significance required for the particular regression model. Verification of statistical significance of the obtained results is confirmed by the four separate test statistics calculated to decide about the appropriateness of the created regression models – the Shapiro-Wilk test, the DurbinWatson statistic, the White test, and the variance inflation factor (Durbin & Watson, 1951; Shapiro & Wilk, 1965; White, 1980). The tests mentioned above also contribute to the other analytical studies to verify the statistical significance of the study outcome (Marsillas et al., 2017; Michnevic, 2016; Milanez, 2020; Okely et al., 2018; Ray et al., 2018; Sala, 2020). The whole analysis is executed in the R statistical environment through the programming language R (R Core Team, 2022) with the additional help of the NbClust package (Charrad et al., 2014), the plm package (Croissant & Millo, 2008; Croissant et al., 2017), and the RColorBrewer package (Neuwirth, 2015). 3. Data Description The data comes from the three following databases – the European Union Statistics on Income and Living Conditions, the European Union Labour Force Survey, and the European System of Integrated Social Protection Statistics. The first one collects the timely and comparable cross-sectional and longitudinal multidimensional microdata on income, poverty, social exclusion and living conditions. The second one is the extensive household sample survey that provides the data on labour participation of the population aged 15 and over and on the inactive individuals in the labour market. The third one enables the comparison of the national administrative data on social protection at the international level. It provides a coherent comparison between the European countries of the social benefits to households and their financing. The analysis covers two dimensions: a territorial angle of a view involving the 27 countries of the European Union with an omission of the United Kingdom, and the time perspective represented by the period from 2009 to 2019. The data are composed annually. An observed set of the area involved in the analysis consists of the following countries: Austria (AT), Belgium (BE), Bulgaria (BG), Croatia (HR), Cyprus (CY), Czech Republic (CZ), Denmark (DK), Estonia (EE), Finland (FI), France (FR), Germany (DE), Greece (GR), Hungary (HU), Ireland (IE), Italy (IT), Latvia (LV), Lithuania (LT), Luxembourg (LU), Malta (MT), the Netherlands (NL), Poland (PL), Portugal (PT), Romania (RO), Slovakia (SK), Slovenia (SI), Spain (ES), and Sweden (SE) (International Organization for Standardization, 2013). The United Kingdom is omitted because the future agenda and the policies supplied by the European Union do not cover this former member. The following indicators determining the ageing society, which are related to people aged 55 years and over, are observed in the analysis: old-age dependency ratio (OADR), elderly employment (EE), part-time employment (PTE), weekly work hours (WWH), life expectancy (LE), disability-free life expectancy (DFLE), personal health (PH), health problem limitation (HPL), net income (NI), social protection expenditure (SPE), income inequality (II), at-risk-of-poverty population (PP), and material and social deprivation (MASD). The following variables represent the involved indicators:  Old-age dependency ratio (tps00198): the ratio of the population aged 65+ and the population aged 15 to 64 years expressed per 100 people of the working age from 15 to 64 years (Eurostat, 2021n);  Elderly employment is designated as the employment rate of older workers (tesem050): a percentage of employees aged E+M_01_2023.indb 49 28.2.2023 10:02:35 50 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics 55–64 years in the total population of the same age group (Eurostat, 2021b);  Elderly part-time employment designated as persons employed part-time (tps00159): a percentage of part-time employees aged 55–64 years in the total employment of the same age group (Eurostat, 2021c);  Weekly work hours designated as the average number of usual weekly hours of work in the main job (lfst_r_lfe2ehour): average weekly work hours per employee (Eurostat, 2021d);  Life expectancy designated as life expectancy at age 65 (tps00026): the average number of years remaining to be lived by a person who is aged 65 years (Eurostat, 2021e);  Disability-free life expectancy designated as disability-free life expectancy at 65 (tepsr_sp320): the number of years that a person aged 65 is still expected to live in a healthy condition (Eurostat, 2021f);  Personal health is designated as selfperceived health (hlth_silc_10): the subjectively perceived health by the population aged 55+ years who report the level of their health status as good or very good (Eurostat, 2021g);  Health problem limitation designated as self-perceived long-standing limitations in usual activities due to health problem (hlth_silc_12): the subjective perception of limitations due to health problems by people aged 55+ years who report the level of activity limitation as some or severe (Eurostat, 2021h);  Net income designated as mean equalised net income (ilc_di3): the average net income regarding social transfers and including pensions in purchasing power standard (Eurostat, 2021i);  Social protection expenditure designated as expenditure on social protection per inhabitant (tps00100): the expenditures on social benefits, administration costs and other expenditures measures in purchasing power standard per inhabitant (Eurostat, 2021j);  Income inequality is designated as income inequality for older people (tespn080): the ratio of total equalised disposable income received by the 20% of the population aged 65+ years with the highest income and the 20% of the population aged 65+ years with the lowest income meaning comparing the top and the lowest quintile (Eurostat, 2021k);  At-risk-of-poverty population or social exclusion (ilc_peps01): the at-risk-of-poverty population aged 55+ years or severely materially deprived whose equalised disposable income is below 60% of the national median after social transfers (Eurostat, 2021l);  Material and social deprivation rate (ilc_ mdsd07): the rate of people aged 55+ years who feel materially or socially deprived and they experience at least 4 out of 9 following deprivations items, which cannot afford to pay rent or utility bills, keep home adequately warm, face unexpected expenses, eat meat, fish or a protein equivalent every second day, a week holiday away from home, a car, a washing machine, a colour television or a telephone belong among (Eurostat, 2021m). The indicators applied in the analysis are also involved in many other studies regarding the ageing society. The most common are those related to the participation of the older employees in the labour market as turning the elderly into an active workforce seems like an opportunity to face the challenges of an ageing population (Been & Vliet, 2017; Laun, 2017; Soong, 2020). Then, the indicators that assess the financial security of the elderly are proven to be significant both in the analysis and the other researches (Antonelli & Bonis, 2019; Guido et al., 2020; Heuvel & Olaroiu, 2017). Their mutual relations and correspondence represent a supportive element of the analytical outcome. Because ageing is a multidimensional process, the main focus is on the so-called technical aspect of the ageing society as there are many other views on this issue. 4. Research Results, Discussion, and Limitations The analytical process consists of the two main phases – the cluster analysis that distinguishes the explored countries into the clusters and regression analysis to construct the panel regression models assigned to the particular clusters. 4.1 Cluster Analysis Several clusters are qualified applying the selected approaches – the Ball-Hall index, the McClain-Rao index, and the point-biserial correlation coefficient. The number of clusters E+M_01_2023.indb 50 28.2.2023 10:02:35 51 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics representing the ageing society indicators among the observed countries is set up to three (Tab. 1). The list of most similar and the most dissimilar pairs of countries according to the Euclidean distance values of all the possible pairs of the European Union member countries is summarised in Tab. 2. The six countries create the most similar pairs (Spain, Italy, Latvia, Lithuania, Denmark, and Finland). The most extreme pairs are represented by Spain and Italy. The most similar pair mutually, the nearest one of the explored spans, are created by Latvia and Lithuania in 2012. The five countries represent the most different countries – Netherlands, Bulgaria, Sweden, Latvia, and Lithuania. The most considerable disproportion during the whole observed period is found between the Netherlands and Bulgaria in 2010. Latvia and Lithuania appear on both sides as they created the nearest pair and the outermost pair several times. The similarity of Italy and Spain in the case of the ageing population is observed since 2009 in the analysis. By 2019, these two countries are characterised by the most significant similarity among all the European Union member countries eight times. In recent years, several authors have tested the differences and the similarities between these two countries. Tomassini and Lamura (2009) confirm that Italy and Spain are the countries where the proportion of the older population, the median age of the population, and the ageing index achieve the highest values. The other studies point to the relatively similar informal care from outside the household, while their long-term care public benefits systems are different. Also, these two countries have equal ratios of long-term care Method Test statistic Test statistic value Clusters Ball-Hall index Barycentre mean dispersion 268.8438 3 McClain-Rao index Denominator 0.6965 3 Point-biserial coefficient Correlation 0.6613 3 Source: own Year Nearest pair of countries Outermost pair of the country Distance Country 1 Country 2 Distance Country 1 Country 2 2009 1.49230 ES IT 9.02643 NL BG 2010 1.62560 LV LT 9.34777 NL BG 2011 1.85518 ES IT 9.05030 NL BG 2012 1.18115 LV LT 9.14514 SE BG 2013 1.32540 ES IT 9.20414 NL BG 2014 1.22483 ES IT 8.67009 LV SE 2015 1.20303 ES IT 8.85431 NL BG 2016 1.42708 ES IT 8.55801 NL BG 2017 1.58530 DK FI 8.55027 LV SE 2018 1.34759 ES IT 8.52485 LV SE 2019 1.21543 ES IT 8.58197 LT SE Source: own Tab. 1: The numbers of clusters of the observed countries according to an ageing society Tab. 2: The most similar and the most dissimilar countries according to an ageing society E+M_01_2023.indb 51 28.2.2023 10:02:36 52 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics public spending as a percentage of the gross domestic product (Courbage et al., 2020). Moreover, the diversity of the Netherlands and Bulgaria is also confirmed. When clustering the European Union member countries according to the indicators based on the active ageing index, the Netherlands are involved in the cluster with the highest performance, while Bulgaria is assigned to the third cluster with the countries that achieve the lower performance (Thalassinos et al., 2019). The conditions at the beginning of the analysed period are shown in Fig. 1. The first cluster consists of countries with similar historical backgrounds, namely the Baltic countries, the Visegrad Group countries, and the SouthEastern European countries. A similar situation appears in the second cluster, created by the Mediterranean countries. The third cluster contains mainly the Benelux countries and the Scandinavian countries. When comparing to the end of the analysed period in 2019, changes in the distribution of clusters are visible in Fig. 2. Homogeneity of the created clusters has disappeared throughout the observed period. Only the Visegrad Group countries have remained in the first cluster. It is probably related to these countries joining the European Union considerably later than the remaining countries in 2004. Assignment of the other countries has dispersed all over the European Union area. The critical dissimilarity is found at the end of the examined period, and hence, the influence of the original grouping of the European Union member countries plays a more important role here. The summarising scenario is created according to the mean Euclidean distances between the individual countries during the observed period, as shown in Fig. 3. It clarifies the classification of the countries, which are similar in an area of ageing society during the whole observed period. The results of the summarising view underline the importance of the original geographical grouping of the European countries and their neighbourhood. Also, there is a group of the transitive economies visibly related mutually in the first cluster. It does not significantly differ from those at the initial or terminated point of the explored time. The following heat maps illustrate the similarity between each possible pair of encompassed countries. Each cell is assigned the shade of grey – the darker colour, the more distant pair of the countries. It means such countries were more similar in ageing society indicators. The first heat map visualised a situation at the beginning of the explored period in Fig. 1: The dendrogram of ageing society similarity according to the explored countries for the year 2009 Source: own E+M_01_2023.indb 52 28.2.2023 10:02:36 53 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics 2009. As it can be seen from the heat map in Fig. 4, there were countries which are in contrast with the other countries – Bulgaria with the average mutual distance to all the other countries (distance 5.40579), Romania (distance 5.01617), and the Netherlands (distance 4.48775). Fig. 2: The dendrogram of ageing society similarity according to the explored countries for the year 2019 Source: own Fig. 3: The dendrogram of ageing society similarity according to the explored countries for the whole observed period Source: own E+M_01_2023.indb 53 28.2.2023 10:02:37 60 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics Conclusions Population ageing is a common experience of all the European Union member countries, and there are several differences among them from a perspective of the involved indicators. These are noticeable whether they are the health indicators, the labour conditions for the elderly population, or the financial security for pension in each country. This paper investigates the ageing process in the European Union and the differences and similarities between the selected countries. The similarity of the countries is assessed through seven statistically significant indicators belonging to the characteristics of the society. The focus is on the age group of 55+ as it represents the elderly population group that is disadvantaged from a perspective of age. The quantitative methods are applied using cluster analysis and sensitivity analysis. It was found that the most similar pairs of the countries during the observed period were Latvia with Lithuania and Italy with Spain. On the contrary, the most considerable disparities were found in the case of Bulgaria and the Netherlands. The Visegrad Group member countries and the Baltic countries keep their positions in the first cluster during the whole explored period. Also, the Scandinavian countries were localised in the third cluster for the whole explored period. It suggests that similar trends within the population ageing have long been presented in the given geographical areas. The panel regression model also demonstrates that not all the variables are statistically significant in the individual cluster of panel data regression models. The statistically significant ones are the part-time employment, the life expectancy, the income inequality, the weekly work hours, the net income, the material and social deprivation, and the at-risk-of-poverty population. They are all assigned to the age group of 55 years of age and over. Only the three of them, which the part-time employment, the income inequality, and the material and social deprivation belong among, are significant in all the regression models for the whole area and the three clusters. The obtained results could serve as a resource for the research and the national policies focused on the quality of life and the living conditions of the elderly population as it can help them to identify the objectionable determinants and to implement the appropriate measures in order to improve the situation and position of this population group. Acknowledgements: Supported by the Scientific Grant Agency of the Ministry of Education, Science, Research, and Sport of the Slovak Republic and the Slovak Academy of Sciences within the project “VEGA 1/0646/23”. References Aguilar-Palacio, I., Gil-Lacruz, A. I., Sánchez-Recio, R., & Rabanaque, M. J. (2018). Self-rated health in Europe and its determinants: Does generation matter? International Journal of Public Health, 63, 223–232. https:// doi.org/10.1007/s00038-018-1079-5 Antonelli, M. A., & Bonis, V. (2019). The efficiency of social public expenditure in European countries: A two-stage analysis. Applied Economics, 51(1), 47–60. https://doi.org/10.10 80/00036846.2018.1489522 Ball, G. H., & Hall, D. J. (1965). Isodata: A novel method of data analysis and pattern classification. Stanford Research Institute. Bates, D., Chambers., J., Dalgaard, P., Gentleman, R., Hornik, K., Ihaka, R., Kalibera, T., Lawrence, M., Leisch, F., Ligges, U., Lumley, T., Maechler, M., Meyer, S., Murrell, P., Plummer, M., Ripley, B., Sarkar, D., Lang, D. T., Tierney, L., Urbanek, S., Schwarte, H., Masarotto, G., Iacus, S., Falcon, S., Murdoch, D., & Morgan, M. (2022). R: Language and environment for statistical computing. R Foundation for Statistical Computing. Retrieved June 22, 2022, from https://www.R-project.org Been, J., & Vliet, O. (2017). Early retirement across Europe. Does non‐standard employment increase participation of older workers? International Review for Social Sciences, 70(2), 163–188. https://doi.org/10.1111/kykl.12134 Best, R., & Hill, P. (2020). What is the purpose of retiring? Innovation in Aging, 4(1), 442. https://doi.org/10.1093/geroni/igaa057.1430 Bornet, M., Truhard, E. R., Rochat, E., Pasquier, J., & Monod, S. (2017). Factors associated with quality of life in elderly hospitalized patients undergoing post-acute rehabilitation: A cross-sectional analytical study in Switzerland. BMJ Open, 7, 1–8. https://doi.org/10.1136/ bmjopen-2017-018600 Broniatowska, P. (2019). Population ageing and inflation. Journal of Population Ageing, 12, 179–193. https://doi.org/10.1007/ s12062-017-9209-z Brønnum-Hansen, H., Foverskov, E., & Andersen, I. (2021). Income inequality in E+M_01_2023.indb 60 28.2.2023 10:02:40 61 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics life expectancy and disability-free life expectancy in Denmark. Journal of Epidemiology and Community Health, 75, 145–150. https://doi. org/10.1136/jech-2020-214108 Charrad, M., Ghazzali, N., Boiteau, V., & Niknafs, A. (2014). NbClust: An R package for determining the relevant number of clusters in a data set. Journal of Statistical Software, 61(6), 1–36. https://doi.org/10.18637/jss.v061.i06 Coleman, D., & Rowthorn, R. (2011). Who’s afraid of population decline? A critical examination of its consequences. Population and Development Review, 37(s1), 217–248. https:// doi.org/10.1111/j.1728-4457.2011.00385.x Courbage, C. H., Montoliu-Montes, G., & Wagner, J. (2020). The effect of long-term care public benefits and insurance on informal care from outside the household: Empirical evidence from Italy and Spain. The European Journal of Health Economics, 21, 1131–1147. https://doi.org/10.1007/s10198-020-01215-7 Cristea, M., Noja, G. G., Stefea, P., & Sala, A. L. (2020). The impact of population aging and public health support on EU labor markets. International Journal of Environmental Research and Public Health, 17(4), 1439–1466. https://doi.org/10.3390/ijerph17041439 Croissant, Y., & Millo, G. (2008). Panel data econometrics in R: The plm package. Journal of Statistical Software, 27(2), 1–43. https://doi. org/10.18637/jss.v027.i02 Croissant, Y., Millo, G., Tappe, K., Toomet, O., Kleiber, C., Zeileis, A., Henningsen, A., Andronic, L., & Schoenfelder, N. (2017). Package plm. R Foundation for Statistical Computing. Retrieved June 22, 2022, from https:// cran.r-project.org/web/packages/plm/plm.pdf Durbin, J., & Watson, G. S. (1951). Testing for serial correlation in the least squares regression. II. Biometrika, 38(1–2), 159–178. https:// doi.org/10.1093/biomet/38.1-2.159 Ebbinghaus, B. (2021). Inequalities and poverty risks in old age across Europe: The double‐edged income effect of pension systems. Social Policy & Administration, 55(3), 440–455. https://doi.org/10.1111/spol.12683 Estevens, J. (2018). Migration crisis in the EU: Developing a framework for analysis of national security and defence strategies. Comparative Migration Studies, 6, 28. https:// doi.org/10.1186/s40878-018-0093-3 Eurostat. (2021a). Closing the gender pay gap? European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa. eu/eurostat/en/web/products-eurostat-news/-/ DDN-20210203-1 Eurostat. (2021b). Employment rate of older workers, age group 55–64. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/ view/tesem050/default/table?lang=en Eurostat. (2021c). Persons employed part-time: Total. European Commission – Eurostat. Retrieved June 22, 2022, from https:// ec.europa.eu/eurostat/databrowser/view/ tps00159/default/table?lang=en Eurostat. (2021d). Average number of usual weekly hours of work in the main job by sex, age and NUTS 2 regions (hours). European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/ view/lfst_r_lfe2ehour/default/table?lang=en Eurostat. (2021e). Life expectancy at 65, by sex. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa. eu/eurostat/databrowser/view/tps00026/ default/table?lang=en Eurostat. (2021f). Healthy life years at age 65 by sex. European Commission – Eurostat. Retrieved June 22, 2022, from https:// ec.europa.eu/eurostat/databrowser/view/ tepsr_sp320/default/table?lang=en Eurostat. (2021g). Self-perceived health by sex, age and income quintile. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/ view/hlth_silc_10/default/table?lang=en Eurostat. (2021h). Self-perceived longstanding limitations in usual activities due to health problem by sex, age and income quintile. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/view/hlth_silc_12/default/ table?lang=en Eurostat. (2021i). Mean and median income before social transfers (pensions included in social transfers) by age and sex. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/view/ILC_DI13__custom_545861/ default/table?lang=en Eurostat. (2021j). Expenditure on social protection per inhabitant. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/ view/tps00100/default/table?lang=en Eurostat. (2021k). Income inequality for older people. European Commission E+M_01_2023.indb 61 28.2.2023 10:02:41 62 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/ view/tespn080/default/table?lang=en Eurostat. (2021l). People at risk of poverty or social exclusion by age and sex – EU 2020 strategy. European Commission – Eurostat. Retrieved June 22, 2022, from https:// ec.europa.eu/eurostat/databrowser/view/ ilc_peps01$DV_565/default/table?lang=en Eurostat. (2021m). Material and social deprivation rate by age and sex. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa.eu/eurostat/databrowser/ view/ilc_mdsd07/default/table?lang=en Eurostat. (2021n). Old-age-dependency ratio. European Commission – Eurostat. Retrieved June 22, 2022, from https://ec.europa. eu/eurostat/databrowser/view/tps00198/ default/table?lang=en Fukuda, S., & Okumura, K. (2020). Regional convergence under declining population: The case of Japan. Japan and the World Economy, 55, 101023. https://doi.org/10.1016/j. japwor.2020.101023 Giang, L. T., Nguyen, T. T., & Nguyen, N. T. (2020). Social support and self-rated health among older men and women in Vietnam. Journal of Population Ageing, 13, 427–442. https:// doi.org/10.1007/s12062-020-09283-6 Gómez-Costilla, P., García-Prieto, C., & Somarriba-Arechavala, N. (2021). Aging and gender health gap: A multilevel analysis for 17 European countries. Social Indicators Research, (160), 1051–1069. https://doi. org/10.1007/s11205-020-02595-2 Grannis, S. J., Xu, H., Vest, J. R., Kasthurirathne, S., Bo, N., Moscovitch, B., Torkzadeh, R., & Rising, J. (2019). Evaluating the effect of data standardization and validation on patient matching accuracy. Journal of the American Medical Informatics Association, 26(5), 447–456. https://doi.org/10.1093/jamia/ ocy191 Guido, G., Amatulli, C., & Sestino, A. (2020). Elderly consumers and financial choices: A systematic review. Journal of Financial Services Marketing, 25, 76–85. https://doi.org/10.1057/ s41264-020-00077-7 Heuvel, W., & Olaroiu, M. (2017). How important are health care expenditures for life expectancy? A comparative, European analysis. Journal of the American Medical Directors Association, 18(3), 276.e9-276.e12. https://doi. org/10.1016/j.jamda.2016.11.027 International Organization for Standardization. (2013). International standard for country codes and codes for their subdivisions. International Organization for Standardization 3166. Retrieved June 22, 2022, from https://www.iso. org/obp/ui/#iso:std:iso:3166:-1:ed-3:v1:en,fr Jakovljevic, M. M., Netz, Y., Buttigieg, S. C., Adany, R., Laaser, U., & Varjacic, M. (2018). Population ageing and migration-history and UN forecasts in the EU-28 and its east and south near neighbourhood-one century perspective 1950–2050. Globalization and Health, 14, 30. https://doi.org/10.1186/s12992-018-0348-7 Janus, J., Bożek, P., Taszakowski, J., & Doroż, A. (2022). Decaying villages in the centre of Europe with no population decline: Long-term analysis using historical aerial images and remote sensing data. Habitat International, 121, 102520. https://doi.org/10.1016/j. habitatint.2022.102520 Khan, H. (2019). Population ageing in a globalized world: Risk and dilemmas? Journal of Evaluation in Clinical Practice, 25(5), 754–760. https://doi.org/10.1111/jep.13071 Kikuchi, H., Emberger, G., Ishida, H., Fukuda, A., & Kobayakawa, S. (2022). Dynamic simulations of compact city development to counter future population decline. Cities, 127, 103753. https://doi.org/10.1016/j.cities.2022.103753 Kluge, F. A., Goldstein, J. R., & Vogt, T. C. (2019). Transfers in an ageing European Union. The Journal of the Economics of Ageing, 13, 45–54. https://doi.org/10.1016/j.jeoa.2018.07.004 Kratt, O., & Kirnos, I. (2020). Cross-national analysis of the older workers’ employment rate. SHS Web of Conferences, 73, 01014. https:// doi.org/10.1051/shsconf/20207301014 Laun, L. (2017). The effect of age-targeted tax credits on labour force participation of older workers. Journal of Public Economics, 152, 102–118. https://doi.org/10.1016/j. jpubeco.2017.06.005 Le, D. D., Leon-Gonzalez, R., & Giang, T. L. (2020). Decomposing gender inequality in functional disability among older people in Vietnam. Archives of Gerontology and Geriatrics, 87, 56–68. https://doi.org/10.1016/j. archger.2019.103989 Leeuwen, K., Loon, M., Nes, F., Bosmans, J., Vet, H., Ket, J., Widdershoven, G., & Ostelo, R. (2019). What does quality of life mean to older adults? A thematic synthesis. PLoS ONE, 14(3), 1–39. https://doi. org/10.1371/journal.pone.0213263 E+M_01_2023.indb 62 28.2.2023 10:02:41 63 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics Li, L., Du, T., & Hu, Y. (2020). The effect of population aging on healthcare expenditure from a healthcare demand perspective among different age groups: Evidence from Beijing city in the People’s Republic of China. Risk Management and Healthcare Policy, 13, 1403– 1412. https://doi.org/10.2147/RMHP.S271289 Liu, J., Wei, W., Peng, Q., & Xue, C. (2019). Perceived health and life satisfaction of elderly people: Testing the moderating effects of social support, attitudes toward aging, and senior privilege. Journal of Geriatric Psychiatry and Neurology, 33(3), 144–154. https://doi. org/10.1177/0891988719866926 Loichinger, E., & Pothisiri, W. (2018). Health prospects of older persons in Thailand: The role of education. Asian Population Studies, 14(3), 310-329. https://doi.org/10.1080/17441730.201 8.1532140 Lutz, W., Amran, G., Belanger, A., Conte, A., Gailey, N., Ghio, D., Grapsa, E., Jensen, K., Loichinger, E., Marois, G., Muttarak, R., Potancokova, M., Sabourin, P., & Stonawski, M. (2019). Demographic scenarios for the EU: Migration, population and education. Publications Office of the European Union. Lyons, A. C., Grable, J. E., & Joo, S.-H. (2018). A cross-country analysis of population ageing and financial security. The Journal of the Economics of Ageing, 12, 96–117. https:// doi.org/10.1016/j.jeoa.2018.03.001 Marois, G., Bélanger, A., & Lutz, W. (2020). Population aging, migration, and productivity in Europe. Proceedings of the National Academy of Sciences, 117(14), 7690–7695. https://doi. org/10.1073/pnas.1918988117 Marsillas, S., De Donder, L., Kardol, T., Regenmortel, S., Dury, S., Brosens, D., Smetcoren, A.-S., Brana, T., & Varela, J. (2017). Does active ageing contribute to life satisfaction for older people? Testing a new model of active ageing. European Journal of Ageing, 14, 295–310. https://doi.org/10.1007/ s10433-017-0413-8 McClain, J. O., & Rao, V. R. (1975). Clustisz: A program to test for the quality of clustering of a set of objects. Journal of Marketing Research, 12(4), 456–460. https://www.jstor. org/stable/3151097 Michnevic, K. (2016). The effects of ageing on household consumption in Central and Eastern Europe. Economy & Business, 10, 273–287. Milanez, A. (2020). Workforce ageing and labour productivity dynamics. Our Economy, 66(3), 1–13. https://doi.org/10.2478/ngoe-20200013 Milligan, G. W. (1981). A Monte Carlo study of thirty internal criterion measures for cluster analysis. Psychometrika, 46, 187–199. https:// doi.org/10.1007/BF02293899 Moen, P. (2020). Is working longer the solution to an ageing society? Innovation in Aging, 4(S1), 809–810. https://doi.org/10.1093/geroni/ igaa057.2937 Mura, L., Gontkovicova, B., Dulova Spisakova, E., & Hajduova, Z. (2019). Position of employee benefits in remuneration structure. Transformations in Business & Economics, 18(2), 156–173. Neuwirth, E. (2015). Package “RColorBrewer”. Retrieved June 22, 2022, from https:// cran.r-project.org/web/packages/RColorBrewer/RColorBrewer.pdf Organisation for Economic Co-operation and Development. (2019). Fiscal challenges and inclusive growth in ageing societies. Organisation for Economic Co-operation and Development Economic Policy Papers, 27, 1–70. https://doi.org/10.1787/c553d8d2-en Officer, A., Thiyagarajan, J. A., Schneiders, M. L., Nash, P., & Fuente-Núñez, V. (2020). Ageism, healthy life expectancy and population ageing: How are they related? International Journal of Environmental Research and Public Health, 17(9), 3159–3170. https:// doi.org/10.3390/ijerph17093159 Okely, J., Weiss, A., & Gale, C. (2018). The interaction between individualism and wellbeing in predicting mortality: Survey of health ageing and retirement in Europe. Journal of Behavioral Medicine, 41, 1–11. https://doi.org/10.1007/ s10865-017-9871-x Ortega, J. (2020). Is active ageing coping with population ageing? Journal of Population Ageing, 14, 37–52. https://doi.org/10.1007/ s12062-020-09265-8 Park, C.-Y., Shin, K., & Kikkawa, A. (2022). Demographic change, technological advance, and growth: A cross-country analysis. Economic Modelling, 108, 105742. https://doi. org/10.1016/j.econmod.2021.105742 Pascual-Saez, M., Cantarero-Prieto, D., & Pires Manso, J. R. (2020). Does population ageing affect savings in Europe? Journal of Policy Modelling, 42(2), 291–306. https://doi. org/10.1016/j.jpolmod.2019.07.009 Quattrociocchi, L., Tibaldi, M., Marsili, M., Fenga, L., & Caputi, M. (2020). Active ageing E+M_01_2023.indb 63 28.2.2023 10:02:41 64 2023, volume 26, issue 1, pp. 45–64, DOI: 10.15240/tul/001/2023-1-003 Economics and living condition of older persons across Italian regions. Journal of Population Ageing, 14, 91–136. https://doi.org/10.1007/ s12062-020-09309-z Ray, J., Popli, G., & Fell, G. (2018). Association of cognition and age-related hearing impairment in the English longitudinal study of ageing. JAMA Otolaryngol Head & Neck Surgery, 144(10), 876–882. https://doi.org/10.1001/ jamaoto.2018.1656 R Core Team. (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.Rproject.org/ Rupeika-Apoga, R., & Thalassinos, Y. E. (2019). The impact of population ageing and social stratification: The case of Latvia. International Journal of Economics & Business Administration, 7(1), 49–63. https://doi.org/10.35808/ ijeba/195 Sala, L. A. (2020). The relationship between population ageing and medical expenditures in Romania. Facta Universitatis, 17(2), 157–172. https://doi.org/10.22190/FUEO191223012S Sanderson, W. C., & Scherbov, S. (2015). Are we overly dependent on conventional dependency ratios? Population and Development Review, 41(4), 687–708. https:// doi.org/10.1111/j.1728-4457.2015.00091.x Sanderson, W. C., & Scherbov, S. (2016). A new perspective on patterns of aging in Europe by education and gender. Journal of Population Ageing, 9, 207–225. https://doi. org/10.1007/s12062-015-9125-z Sewdas, R., Wind, A., Zwaan, L., Borg, W., Steenbeek, R., Beek, A., & Boot, C. (2017). Why older workers work beyond the retirement age: A qualitative study. BMC Public Health, 17, 672. https://doi.org/10.1186/s12889-017-4675-z Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3–4), 591–611. https://doi.org/10.1093/biomet/52.3-4.591 Skibiński, A. (2017). The changes of the population structure and its consequences in selected EU countries: Some aspects. European Journal of Sustainable Development, 6(1), 357–368. https://doi.org/10.14207/ejsd.2017. v6n1p357 Soong, J. J. (2020). Empowering the elderly to promote active ageing in the labour market: A new strategic scheme to improve ageing human resource and to solve shortage of labour force in Taiwan. Jebat: Malaysian Journal of History, Politics & Strategic Studies, 47(1), 198–224. Striessnig, E., & Lutz, W. (2013). Can below-replacement fertility be desirable? Empirica, 40, 409–425. https://doi.org/10.1007/ s10663-013-9213-3 Thalassinos, E. I., Cristea, M., & Noja, G. G. (2019). Measuring active ageing within the European Union. Equilibrium, 14(4), 591–609. https://doi.org/10.24136/eq.2019.028 Tomassini, C., & Lamura, G. (2009). Population ageing in Italy and Southern Europe. International Handbook of Population Aging, 1, 69–89. https://doi.org/10.1007/978-1-4020-8356-3_4 United Nations. (2019a). Active ageing index analytical report. Retrieved June 22, 2022, from https://unece.org/population/publications/ active-ageing-index-analytical-report United Nations. (2019b). World population ageing 2019: Highlights. Department of Economic and Social Affairs, Population Division. https:// un.org/en/development/desa/population/ publications/pdf/ageing/WorldPopulationAgeing2019-Highlights.pdf White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica, 48(4), 817–838. https://doi.org/10.2307/1912934 Williamson, B., & Piattoeva, N. (2018). Objectivity as standardization in data-scientific education policy, technology and governance. Learning, Media and Technology, 44(1), 64–76. https://doi.org/10.1080/17439884.2018.1556215 World Health Organization. (2020). Decade of healthy ageing: Baseline report. https://who. int/publications-detail-redirect/9789240017900 E+M_01_2023.indb 64 28.2.2023 10:02:41