Determinants of ageing in Romania: Evidence from regional-level panel analysis

Jemna, Dănuţ-Vasile; David, Mihaela Article Determinants of ageing in Romania: Evidence from regional-level panel analysis CES Working Papers Provided in Cooperation with: Centre for European Studies, Alexandru Ioan Cuza University Suggested Citation: Jemna, Dănuţ-Vasile; David, Mihaela (2021) : Determinants of ageing in Romania: Evidence from regional-level panel analysis, CES Working Papers, ISSN 2067-7693, Alexandru Ioan Cuza University of Iasi, Centre for European Studies, Iasi, Vol. 13, Iss. 1, pp. 17-32 This Version is available at: https://hdl.handle.net/10419/286644 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ CES Working Papers – Volume XIII, Issue 1 This work is licensed under a Creative Commons Attribution License 17 Determinants of ageing in Romania. Evidence from regional-level panel analysis Dănuț-Vasile JEMNA*, Mihaela DAVID** Abstract At different stages, all countries of the world are going through the demographic ageing process, with the most diverse economic and social implications. Romania has entered this process since the communist period, as a result of both the demographic transition and the transformations imposed by the demographic and economic policies implemented by the state. After 1990, the transition period has accelerated the increasing pace of this phenomenon, especially through increased migration, declining fertility and changes in the labour market. The regional economic, social and demographic gaps inherited from the past were accentuated after the fall of the totalitarian regime, so that the population ageing has intensified and it is experienced differently at the territorial level. This study aims at identifying the determinants of the increase in population ageing, along with the decrease in the share of young population, across the eight development regions of Romania, during the 1995- 2018 period. Keywords: demographic transition, demographic ageing, Romania, regional level, heterogeneous panel data Introduction Population ageing is a global phenomenon and affects, to a greater or lesser extent, all countries in the world, raising a series of fundamental questions regarding the future evolution of human society. Within the context of increasing life expectancy and the tendency to prolong the active life of people aged 65 or older, governments and national and international organizations are involved in the development of strategies and policies on the elderly population. According to the UN, the level of global ageing population will increase significantly in the coming decades, from 6% in 1990 to 16% in 2050 (UN, 2019). The main challenges raised by the continuously growing share of older persons in the total population, along with the decrease in the share of young people, are related to ensure healthy lives and to promote well-being for people aged 65 or older, especially by means of the national health and social security system. * Dănuț-Vasile JEMNA is professor at Faculty of Economics and Business Administration, “Alexandru Ioan Cuza” University of Iasi, Romania, e-mail: danu[email protected]. ** Mihaela DAVID is postdoctoral researcher at Doctoral School of Economics and Business Administration, “Alexandru Ioan Cuza” University of Iasi, Romania, e-mail: [email protected]. CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Dănuț-Vasile JEMNA, Mihaela DAVID 18 Empirical studies have shown that the main factors responsible for population ageing are declining fertility, rise of life expectancy and emigration (Hoff, 2011; Bloom and Luca, 2016; Murphy, 2017; Yang et al., 2019; Nagarajan et al., 2020). Demographic ageing refers to the simultaneous phenomenon of increasing the share of elderly population and the one decreasing the share of young people at the level of a country or region. At the same time, the increase in life expectancy leads to an increase in the median age and the share of total elderly population. In turn, migration brings important changes in the age structure of the population. Basically, where net migration is negative, the population will have an aging trend, as young people are more exposed to migration, especially on the ground of continuing their studies and finding work. In 2019, 20.3% of the EU population was aged 65 and over, and by 2050, older persons are expected to account for 28.5 % of the total population. In comparison, the share of older Romanians was lower in 2019 (18.5%). However, the prospects on the evolution of population ageing in Romania are not more encouraging. According to the UN estimates, by 2050, the share of people aged 65 and older in the population will exceed 30%. In Romania, the ageing process of population begun in the communist period, especially after 1980. The share of elderly in the population exceeds that of young people starting with 2009, namely 16% compared to 15%. The phenomenon is becoming more and more alarming especially in rural areas, because this area is the most affected by migration and lower socio-economic conditions. Besides the differences between urban and rural areas, the level of ageing admits important disparities at the level of Romanian counties and regions. The most affected are the counties in the southern Romania, areas with a low degree of urbanization and a low level of economic development (Jemna, 2017, pp. 190-192). This reality will put a lot of pressure on the public pension and health insurance system, on the need to respond to specific problems for this category of population. For the period 1992-2019, the data provided by the Eurostat show an increase in the share of elderly for all regions (Figure 1a), and the demographic ageing index displays not only an intensification of the phenomenon, but also an increase in regional differences (Figure 1b). In this respect, the southern regions of the country (South-West Oltenia and South-Muntenia) stand out with the highest levels of population ageing. The Bucharest-Ilfov region has a growing trend until 2005, followed by a period of stagnation, with slight variations of the old-age dependency ratio, but below the national average for the entire period. It is also interesting to note the evolution of the North-East region, which has a significant increase in ageing index between 1999 and 2008, 16.9% and 20.6% respectively. However, the following period is highlighted by a decline in population ageing, which places it in the most advantageous position over the other regions. CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Determinants of ageing in Romania. Evidence from regional-level panel analysis 19 Figure 1. The evolution of the share of elderly population (a) and of the old-age dependency ratio (b) by regions, 1992-2019 10 11 12 13 14 15 16 17 18 19 96 98 00 02 04 06 08 10 12 14 16 18 BUCURESTI - ILFOV CENTRU NORD-EST NORD-VEST SUD-EST SUD-MUNTENIA SUD-VEST VEST a 14 16 18 20 22 24 26 28 96 98 00 02 04 06 08 10 12 14 16 18 BUCURESTI - ILFOV CENTRU NORD-EST NORD-VEST SUD-EST SUD-MUNTENIA SUD-VEST VEST b Source: Own representations using Eurostat data Empirical studies on the ageing population in Romania, both qualitative (Crăciun, 2011) and quantitative (Gîrleanu-Şoitu, 2004; Precupetu et al., 2019), are relatively few, and those in regional profile are missing. The importance of studies on this topic is underlined by the pace with which the phenomenon develops and, especially, by its implications on society as a whole. Within the context of the analyses and plans developed at the EU level, Romania has adopted a series of strategies on the elderly population, such as The National Strategy on Active Ageing Promotion and Protection of Elderly and the Strategic Plan of Actions 2015-2020. This strategy took into consideration a series of measures and actions regarding three directions: adequate financing of the healthcare system for elderlies; prolonging the active life and participation in the labour market; involving people of this age category in various social activities (Romanian Government, 2015). The implementation of such policies requires specialized studies evaluating the main determinants of the demographic ageing at national level and, especially, indicating the existing inequalities at regional level. Besides, to the best of our knowledge, there were no studies carried out to identify the determinants of population ageing at regional level in Romania. Therefore, the present study addresses this need and covers a gap in the related literature by using different estimation techniques for panel data models encompassing the eight development regions of Romania for the period 1995-2018. The paper is organized as follows. In Section 1, an analysis of the literature is performed to evaluate the results obtained for various countries and socio-economic contexts. Section 2 presents the data used and the methodological strategy. Section 3 presents and discusses the results obtained at regional level. The paper ends with a set of conclusions and references. CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Dănuț-Vasile JEMNA, Mihaela DAVID 20 1. Literature review Demographic ageing refers to the simultaneous phenomenon of increasing the share of the elderly population and decreasing the share of young people in a country or region. The topic is of great interest and raises a series of fundamental questions regarding the evolution of human society in the future. Population ageing is the 21st century’s dominant demographic phenomenon and affects, to a greater or lesser extent, all the countries in the world (UN, 2019). Due to the complexity of the phenomenon, not only demographers, but also economists, sociologists, psychologists, doctors, biologists, etc. put a lot of effort in understating the process of population ageing. Demographers are interested in explaining the change in population structure using theories related to the nature of the evolution of demographic phenomena, while specialists in other fields aim to assess this phenomenon in relation to population health, insurance and social assistance systems, income, labour market, globalization, etc. Interdisciplinary studies focus on identifying both the determinants of population ageing process and its effects on economic and social life. According to the aim of this research, the discussion on related literature is limited to the impact of demographic, economic and social factors on ageing. 1.1. Ageing and demographic factors From a demographic perspective, research on population ageing highlights that demographic transition has inevitably led to a change in population structure by age groups (Robine and Michel, 2004; UN, 2019; Garcia et al., 2019; Harper, 2019). The decrease in fertility and mortality rates, as basic phenomena of the demographic transition, led to the decrease in the share of young people along with an increase in the share of the elderly (Notestein, 1954). Mainly, the decrease in fertility has led to an increase in the median age of the population, as well as to a decrease in the share of the young population, while the population over 65 has also increased (Murphy, 2017; Harper, 2019; Garcia et al., 2019). Moreover, the decline in mortality has had an impact on increasing the degree of ageing (Murphy, 2017), but differentiated from one country to another and from one age group to another. The impact of mortality is more important in the final phase of the transition process, when low mortality rates increase the share of the elderly. Particularly in the developed countries, where even if the fertility level will not change significantly, the decrease in mortality will lead to an increase in the share of elderly population (Casseli and Vallin, 1990). CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Determinants of ageing in Romania. Evidence from regional-level panel analysis 21 Along with declining fertility and mortality, life expectancy is also a strong factor correlated with demographic ageing. Increasing life expectancy is a common phenomenon in all countries and depends on the progress in health and living standards. This evolution has an important contribution to increasing the share of the elderly population (Harper, 2019). Consequently, this led to a special interest in studies analysing the life expectancy of the elderly in relation to the quality of their life. According to the estimates performed by the UN, a 65-year-old person is currently hoping to live at least 17 years (UN, 2019). Regarding the quality of this period of life, an indicator called healthy life expectancy was developed, which measures the average number of years that a person lives in healthy conditions (Sullivan, 1971; WHO, 2016). For the elderly people, the lack of morbidities and disabilities is especially taken into account. It should also be noted that life expectancy is higher in women by almost 5 years, making the female population older than the male one (UN, 2019). Existing research also shows that migration has an important influence on the degree of demographic ageing (Findlay and Wahba, 2013; Neumann, 2013; Garcia et al., 2019; Nagarajan et al., 2020). Emigration tends to aggravate the phenomenon, because young people are the most willing to emigrate, which also affects fertility levels. At country level, ageing can also be explained by the immigration of elderly people who are looking for a quality living environment (Garcia et al., 2019). In return, internal migration can contribute to increasing disparities in the population structure by groups. More developed and urbanized areas are more attracting for young people, while the elderly are less mobile and are more connected to the places where they have already lived their lives (Garcia et al., 2019). Last but not least, besides geographical and sex differences, the ageing process also displays differences in terms of urbanization, which is also an important determinant of demographic transition. Whilst the development of cities attracts young people, the rural area is ageing at a fairly high rate. This is not only due to the attraction of young people, but also because of the immigration of adults who are looking for quieter and safer areas. The rural space is also exposed to continuous transformation and is very heterogeneous, both from a demographic and socio-economic perspective. In the former communist countries, including Romania, the rural-urban imbalances in the population structure are also caused by internal migration as a result of the state’s policies of industrialization and forced urbanization during the 1960-1980 period (Neményi, 2011). CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Dănuț-Vasile JEMNA, Mihaela DAVID 22 1.2. The impact of economic factors on ageing Another special focus is put on studies that take into account the relationship between changing population structure and economic activity. Ageing is usually considered to have an important impact on economic development, but there are also studies that take into account the reverse causality (Alders and Broer, 2004; Nagarajan et al., 2020). The UN reports on the ageing phenomenon show disparities in regional economic development. Thus, trends indicate that the most demographically aged areas are represented by the developed countries in Asia, the USA, Europe, Australia and New Zealand (UN, 2019), while developing countries have entered in an accelerated ageing process. Moreover, quantitative and qualitative empirical studies have shown that the level of income, the degree of development of a region, the level of activity of the elderly population, the employment rate of female population, or the unemployment have a significant impact on the ageing population (Bagheri-Nesami and Shorofi, 2014; Hsu et al., 2019; Nagarajan et al., 2020). Therefore, the impact of economic factors differs from one country to another and from one development region to another. 1.3. Social determinants of population ageing Other studies analyse the relationship between demographic ageing, level of education and public policies in healthcare system. Education is considered one of the most important determinants of demographic transition and, implicitly, of changing the age structure of the population (Canning, 2011; Murtin, 2013). Increasing the level of education in a country has an impact not only on the labor market, fertility and family decisions, but also on mental health and an increased healthy life expectancy for the elderly population (Schneeweis et al., 2014). Demographic changes during the transition period have been accompanied by a development of medical technology and a permanent increase in healthcare and access to health services. All of this has contributed to increasing lifespan, lowering mortality and increasing population ageing. For developing countries, there is a positive impact of policies that have improved the healthcare system on ageing population (Nagarajan et al., 2020). In addition to the improvements in health services, the increase in the share of elderly is also determined by lifestyle changes, especially those related to diet and reducing health risk factors such as tobacco and alcohol consumption (WHO, 2016; UN, 2019). For Romania, the ageing process of population is inevitable, as in all countries in the region. The phenomenon has a series of specificities related to the history of the last 80 years, but also to the existing differences at regional level. The analysis of the factors that contribute to the increase in CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Determinants of ageing in Romania. Evidence from regional-level panel analysis 23 population ageing and of the existing disparities among regions represent an important basis for designing public policies in an era of demographic ageing which recalls for concrete actions to face its implications at the level of the entire society. In this study, based on the existing literature, we assess the extent to which a number of demographic, economic and social factors have an impact on population ageing across the eight development regions of Romania. The phenomenon should not be seen as a burden or a negative aspect of the evolution of society, but it is imperative that policy makers be prepared to face the demographic changes that are taking place and to ensure a long and healthy life for the elderly. 2. Data and methodology 2.1. Data used In order to identify the main determinants of demographic ageing at the level of the eight development regions of Romania, during the 1995-2018 period, we rely on data provided by Eurostat. For the analysis of the ageing trends we resorted to old-age dependency ratio, which is defined as the number of persons aged 65 and over per 100 persons of working age (15-64). In line with the literature on the determinants of population ageing, we consider the following explanatory variables: real GDP per capita expressed in current prices, lei (used to measure the regions’ economic performance); number of tertiary education graduates per 1,000 inhabitants (as a proxy for the human capital variable); female employment rate (defined as the percentage of females aged 15 and above who are active in the labour force); unemployment rate (defined as the number of unemployed persons divided by the labour force, where the labour force is the number of unemployed persons plus the number of employed persons); number of doctors per 1,000 inhabitants (as a proxy to measure the improvements in healthcare services); degree of urbanization (defined as the share of people living in urban areas); net migration rate (expressed as average annual net number of migrants per 1,000 population). Most empirical studies from literature use the data transformed by logarithm to homogenize the data series. Thus, in the econometric modelling step of this study, all variables are used following the transformation with the log operator, with the exception of net migration rate for which negative values are also registered. CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Dănuț-Vasile JEMNA, Mihaela DAVID 24 2.2. Empirical strategy The empirical investigation consists of several steps: (1) test for stationarity; (2) test for cointegration; (3) panel data modelling. Stationarity and panel cointegration tests are employed to avoid the “spurious” regression problem. Considering the specificity of our panel data, we explore the time series properties of variables using three types of panel unit root tests. From the first category, we employed a Fisher-type test developed by Choi (Fisher-ADF, 2001) and the test proposed by Im, Pesaran, and Shin (IPS, 2003) because these approaches allow the autoregressive parameter to be specific to each region and also because they do not impose the restriction of a balanced panel. From the second category, we selected the test introduced by Levin, Lin, and Chu (LLC, 2000) for considering the relatively small size of the panel, which is inherent when analysing data at a regional level, over a short period taken as a reference. Given that variables are integrated of the same order, we test for cointegration, by relying on the cointegration tests developed by Pedroni (2004), Kao (1999), and Maddala and Wu (1999). These tests assume the null hypothesis of no cointegration, against different specifications of the alternative hypothesis of cointegration. The rejection of no cointegration assumption highlights that a long run relationship exists between series. To assess the long run relationship between population ageing and different determinants, in this paper we build several panel models with different specifications. In this respect, the difference between pooled and heterogeneous panel data is underlined. Specifically, the pooled time series regression equation is: 𝑙𝑛𝑂𝐴𝐷𝑅𝑖𝑡 =𝛽0+𝛽𝑙𝑛𝑋𝑖𝑡 ′+𝜀𝑖𝑡, 𝑖=1,𝑛      , 𝑡=1,𝑇      , (1) where 𝑙𝑛𝑂𝐴𝐷𝑅𝑖𝑡 is the logarithm of old-age dependency ratio, 𝑙𝑛𝑋𝑖𝑡 ′ is the vector of independent variables expressed in logarithmic form, and 𝜀𝑖𝑡 is the error term assumed with conditional mean zero and independent of 𝑋𝑖𝑡 ′. The 𝛽 coefficients represent partial elasticities of old-age dependency ratio in relation to each independent variable. According to Wooldridge (2015), the key feature of panel data that distinguishes them from a pooled cross section is that the same cross-sectional units are followed over a given time period. In other words, the pooled OLS estimation is simply a standard OLS technique run on panel data, which assumes that is no heterogeneity, i.e. the specific effects of each cross-sectional unit are completely ignored. CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Determinants of ageing in Romania. Evidence from regional-level panel analysis 31 Hoff, A. (ed.) (2011), Population ageing in Central and Eastern Europe: Societal and policy implications, Surrey: Ashgate Publishing Limited. Hsu, H.C., Liang, J., Luh, D.L., Chen, C.F. and Wang, Y.W. (2019), Social determinants and disparities in active aging among older Taiwanese, International Journal of Environmental Research and Public Health, 16(16), 3005. Im, K.S., Pesaran, M.H. and Shin, Y. (2003), Testing for unit roots in heterogeneous panels, Journal of Econometrics, 115(1), pp. 53-74. Jemna, D.V. (2017), Demografia României, Iaşi: Editura Universităţii „Al.I. Cuza”. Jemna, D.V. and David, M. (2018), Post-transitional regional fertility in Romania, Demographic Research, 38(57), pp. 1733-1776. Kao, C. (1999), Spurious regression and residual based tests for cointegration in panel data, Journal of Econometrics, 90, pp. 1-44. Levin, A., Lin, C.F. and Chu, C. (2002), Unit root test in panel data: Asymptotic and finite sample properties, Journal of Econometrics, 108(1), pp. 1-24. Madala, G.S. and Wu, S. (1999), A comparative study of unit root tests with panel data and a new simple test, Oxford Bulletin of Economics and Statistics, 61, pp. 631-652. Mark, N.C. and Sul, D. (2003), Cointegration vector estimation by panel DOLS and long-run money demand, Oxford Bulletin of Economics and Statistics, 65(5), pp. 0305-9049. Murphy, M.J. (2017), Demographic determinants of population aging in Europe since 1850. Population and Development Review, 43(2), pp. 257-283. Murtin, F. (2013), Long-term determinants of the demographic transition, 1870–2000, The Review of Economics and Statistics, 95, pp. 617-631. Nagarajan, R., Teixeira, A.A.C. and Silva, S.T. (2020), Ageing population: Identifying the determinants of ageing in the least developed countries, Population Research and Policy Review, 42, pp. 251-273. Neményi, Á. (2011), Demographic ageing in Romania: General and specific consequences on the rural population and the relation to international migration, in: Hoff, A. (ed.), Population ageing in Central and Eastern Europe: Societal and policy implications, London: Routledge, pp. 149-167. Neumann, U. (2013), Are my neighbors ageing yet? Local dimensions of demographic change in German cities, Journal of Population Ageing, 6, pp. 189-209. Notestein, F. (1954), Some demographic aspects of aging, Proceedings of the American Philosophical Society, 98(1), pp. 38-45. CES Working Papers | 2021 - volume XIII(1) | wwww.ceswp.uaic.ro | ISSN: 2067 - 7693 | CC BY Dănuț-Vasile JEMNA, Mihaela DAVID 32 Pedroni, P. (1997), Panel cointegration: Asymptotic and finite sample properties of pooled time series tests, with an application to the PPP hypothesis: new results, Working paper, Department of Economics, Indiana University. Pedroni, P. (2001), Fully modified OLS for heterogeneous cointegrated panels, in: Baltagi, B.H., Fomby, T.B. and Carter Hill, R. (eds.), Nonstationary Panels, Panel Cointegration and Dynamic Panels, Advances in Econometrics, 15, Bingley: Emerald Group Publishing Limited, pp. 93-130. Pedroni, P. (2004), Panel cointegration: Asymptotic and finite sample properties of pooled time series tests with an application to the PPP hypothesis, Econometric Theory, 20, pp. 597-625. Phillips, P. and Moon, H. (1999), Linear regression limit theory for nonstationary panel data, Econometrica, 67, pp. 1057-1112. Precupetu, I., Aartsen, M. and Vasile, M. (2019), Social exclusion and mental wellbeing in older Romanians, Social Inclusion, 7(3), pp. 4-16. Robine, J.M. and Michel, J.P. (2004), Looking forward to a general theory on population aging, The Journals of Gerontology: Series A, 59 (6), pp. 590-597. Romanian Government (2015), National Strategy to promote active ageing and protection of the elderly for the period 2015-2020 (retrieved from https://gov.ro/en/government). Schneeweis, N., Skirbekk, V. and Winter-Ebmer, R. (2014), Does education improve cognitive performance four decades after school completion?, Demography, 51, pp. 619-643. Sullivan, D.F. (1971), A single index of mortality and morbidity, HSMHA Health Reports, 86(4), pp. 347-354. United Nations (UN), Department of Economic and Social Affairs, Population Division (2019), World Population Ageing 2019: Highlights (ST/ESA/SER.A/430). Wooldridge, J.M. (2015), Introductory econometrics: A modern approach, Boston: Cengage Learning. World Health Organization (WHO) (2016), World health report on estimated healthy life expectancy (HALE). Yang, L., Zhao, K. and Fan, Z. (2019), Exploring determinants of population ageing in Northeast China: From a socio-economic perspective, International Journal of Environmental Research and Public Health, 16(21), 4265.