The Impact of Migration on Population Ageing in Asia 1990-2020: A Decomposition Analysis Using Prospective Age


The Impact of Migration on Population Ageing in Asia 1990-2020: 
A Decomposition Analysis Using Prospective Age

Markus Dörflinger

Abstract: Population ageing has become a global trend, which unfolds at different 
speeds across world regions and countries. In Asia, there are countries with rapidly 
ageing populations and those that continue to maintain a younger age structure. 
One potential driver of this difference is international migration. In this study, I 
assess the impact of migration on population ageing in Asian countries over the 
period 1990-2020. To do so, I propose a refined decomposition method, applying a 
prospective view on population ageing that accounts for variation in life expectancy. 
Using data from the United Nations World Population Prospects 2022, changes in 
the prospective old-age dependency ratio in 51 countries are decomposed into the 
effects of cohort turnover, deaths, changes in life expectancy and net migration. 
The results reveal that cohort turnover and deaths have had the largest impact on 
changes in the prospective old-age dependency ratio over the last three decades, 
whereas the impact of international migration and changes in life expectancy was 
smaller in all countries. However, in countries with either highly negative or highly 
positive net migration, the effect of migration on the age structure is substantial. 
As migration largely occurs at younger ages, high immigration has decelerated or 
even halted the process of population ageing in countries such as Bahrain, Macao, 
Oman and Singapore. The opposite effect is observed in emigration countries such 
as Armenia, Georgia and Timor-Leste. Hence, the large differences in the current 
level of population ageing across Asian countries can at least partly be attributed to 
international migration in the last decades.

Keywords: Population ageing · Asia · Prospective age · Migration · Decomposition 
analysis

1 Introduction

Population ageing, characterized by increasing shares of older individuals in a 
population, is an inevitable part of the demographic transition (Goldstein 2009) 
and linked to various opportunities and challenges for economies and societies 
worldwide (United Nations Department of Economic and Social Affairs (UN DESA) 

Comparative Population Studies ‒ Demographic Trends Around the Globe
Vol. 50 (2025): 1-40 (Date of release: 27.03.2025)

Federal Institute for Population Research 2025  URL: www.comparativepopulationstudies.de
       DOI: https://doi.org/10.12765/CPoS-2025-02
       URN: urn:nbn:de:bib-cpos-2025-02en2



•    Markus Dörflinger2

2023). However, the timing, speed and extent of population ageing vary across 
countries. This is attributable to different fertility and mortality trajectories as well 
as to migration. A large body of research exists on the impact of these factors on the 
age structure using stationary or stable population models (e.g. Alho 2008; Coale 
1957; Espenshade et al. 1982), counterfactual population projections (e.g. Blanchet 
1989; Coale 1986; Coleman 2008; Espenshade 1994; Fihel et al. 2023; Lee/Zhou 2017; 
Lesthaeghe et al. 1988; Lutz/Scherbov 2007; McDonald/Kippen 2001; Murphy 2021), 
decomposition approaches (e.g. Caselli/Vallin 1990; de Beer et al. 2011; Fernandes 
et al. 2023; Horiuchi 1991; Kashnitsky et al. 2017; Murphy 2017; Preston et al. 1989; 
Preston/Vierboom 2021), cointegration analysis (Santis/Salinari 2023) or the variable-r 
method (Canudas-Romo et al. 2021). 

Although the results of these studies may vary in detail depending on 
methodologies, two main conclusions can be drawn. First, while fertility decline is 
the main driver of population ageing, mortality improvements play a substantial 
role at later phases of the demographic transition. Second, migration can have a 
rejuvenating or an ageing effect on a population, depending on the age structure of 
the migrants and of the resident population. In addition to the direct impact at the 
time of migration, migrants’ fertility indirectly affects the age structure in the long 
term (Sobotka 2008). As shown for Europe, different patterns of population ageing 
across countries and regions can be attributed partly to migration (de Beer et al. 
2011; Ghio et al. 2022; Kashnitsky et al. 2017). Similar conclusions about the role of 
migration in ageing countries are drawn from studies on replacement migration, 
indicating that realistic levels of immigration cannot offset population ageing 
completely but can to some extent slow down the process (e.g. Bijak et al. 2008; 
Billari/Dalla-Zuanna 2011; Craveiro et al. 2019; Huguet 2003; UN DESA 2000).

While many empirical studies on the drivers of population ageing focus on 
European and North American countries, Asia presents a particularly intriguing 
world region for a comparative analysis of the impact of migration on age-structure 
changes. First, the level and speed of population ageing vary substantially across 
Asian countries (Balachandran et al. 2020; Gietel-Basten et al. 2016). Aside from the 
prominent example of Japan, which has one of the oldest populations in the world, 
population ageing is rapidly advancing in several other countries (e.g. South Korea), 
while other countries continue to maintain a relatively young age structure. Life 
expectancy has also developed very differently, with some countries (e.g. Hong 
Kong) having some of the highest life expectancies worldwide (UN DESA 2022b).1 

Second, Asian countries demonstrate quite diverse migration patterns, ranging 
from high immigration (e.g. United Arab Emirates) to high emigration (e.g. Georgia) 
(ibid.). Moreover, migration patterns and policies in Asia have been subject to 
profound changes over time (Ali/Cochrane 2024; de Haas et al. 2018; Oishi 2021). In 
this context, understanding the impact of migration on population ageing is crucial 
to comprehending Asia’s heterogeneous demographic landscape. 

1 When referring to the statistical units provided by the UN simply as “countries”, no opinion on the 
legal status of these countries, territories or areas, its authorities, or the delimitation of its frontiers 
or boundaries is expressed.



The Impact of Migration on Population Ageing in Asia 1990-2020    • 3

Within the large body of research on the drivers of age-structure changes, age 
is predominantly measured as chronological age, i.e. the number of years a person 
has lived.2 Chronological indicators of population ageing, such as the old-age 
dependency or the total support ratio, are based on a fixed threshold to define 
who is considered old, mostly at age 60 or 65. However, as noted by Scherbov and 
Sanderson (2020), chronological age is only one dimension of ageing and many 
characteristics of individuals are ignored when using these conventional measures.

Several new approaches have been elaborated recently to account for additional 
characteristics beyond chronological age, using risk of death (Alvarez/Vaupel 
2023; Zuo et al. 2018), health care needs (Spijker 2023), distribution of ages within a 
population (d’Albis/Collard 2013), physical health (Demuru/Egidi 2016; Muszyńska/
Rau 2012; Sanderson/Scherbov 2010), cognitive functioning (Skirbekk et al. 2012) 
and active life expectancy (Manton et al. 2006). Prospective age, likely the most 
common alternative measure of population ageing, links a person’s age to the 
average remaining life expectancy (Sanderson/Scherbov 2005). The central idea of 
this approach is that a person’s characteristics, such as physical and cognitive health, 
depend more on the expected remaining years of life (prospective age) than on the 
number of years lived (chronological age). Hence, prospective measures may better 
capture many of the economic and social implications of population ageing. For 
instance, medical expenditures are concentrated in the final years of life (Sanderson/
Scherbov 2007). Prospective age becomes particularly relevant as (healthy) life 
expectancy tends to increase over time in most countries (Global Burden of Disease 
Collaborative Network 2020; Rau et al. 2008; Vaupel et al. 2021). Thus, a 65-year-old 
today is on average healthier and lives longer than in 1950. In fact, global remaining 
life expectancy at age 65 has increased from 11.3 years in 1950 to 17.5 years in 
2019 (Scherbov et al. 2022). Considering these life expectancy increases and health 
improvements, prospective measures indicate a slower speed of population ageing 
than chronological measures (Sanderson/Scherbov 2008; UN DESA 2019). 

Life expectancy varies not only over time but also across countries and regions. 
For instance, the characteristics of an average person at a given chronological 
age in a low-mortality country (e.g. Japan) are barely comparable to those of an 
average person at the same chronological age in Afghanistan, where mortality is 
much higher and (remaining) life expectancy much lower. These differences become 
explicitly relevant when defining who is considered “dependent” in a population. 
The widely used fixed old-age threshold of 65 years, which marks an historically 
and contemporarily common retirement age in many countries, is based on the 
assumption that the retired population is entirely dependent on the tax contributions 
of those who work (Gietel-Basten et al. 2015). In many Asian countries, however, 
the coverage of the public pension systems is low (Organisation for Economic Co-
operation and Development (OECD) 2022), and informal employment rates are high 
(International Labour Organization (ILO) 2024). In this context, prospective measures 

2 An exception is the study on replacement migration by Craveiro et al. (2019), which uses prospective 
age.



•    Markus Dörflinger4

appear to be more suitable for a comprehensive understanding of population 
ageing in Asia. However, this new perspective is rarely incorporated into the analysis 
of the drivers of population ageing. 

In this paper, I present an approach to incorporate the prospective age concept 
into a decomposition of age-structure changes. The study aims to assess the drivers 
of population ageing in Asian countries over the period 1990-2020, with a particular 
focus on the role of migration.3

2 Data and methods

2.1 Data

This study uses a) annual population and deaths counts by single age and b) life 
expectancy by single age (both sexes) for the definition of prospective age. These 
data are drawn from the United Nations’ World Population Prospects (WPP) 2022 
estimates (UN DESA 2022b). The UN WPP 2022 estimates are based on population 
and household censuses, vital and population registration systems, surveys and 
other sources (e.g. UNHCR data on refugees) (UN DESA 2022a). Following the UN 
definition of geographic regions, 51 Asian countries are included in the analysis: 5 
from Central Asia, 8 from Eastern Asia, 11 from South-Eastern Asia, 9 from Southern 
Asia and 18 from Western Asia. 

2.2 Indicators of population ageing

The old-age dependency ratio of the (dependent) old-age population relative to the 
working-age population is a common measure of the potential burden of population 
ageing. This ratio is used as the key indicator in this study.4 Conventionally, individuals 
are considered to enter old-age at 65, while the working-age population includes 
the population of age 15 to 64. Therefore, the (chronological) old-age dependency 
ratio is defined as follows:

In contrast to the chronological approach, which uses a constant old-age threshold 
at age 65, the prospective old-age threshold (POAT) is determined by the average 

3 In this study, when referring to migration, international migration is implied unless otherwise 
specified.

4 Other studies on population ageing use the total support ratio or potential support ratio, which 
is the reciprocal of the old-age dependency ratio. However, the use of the term “support ratio” is 
sometimes ambiguous (Kashnitsky et al. 2017), so I use the old-age dependency ratio in this study.

(Chronological) old-age dependency ratio (OADR) = old-age population
working-age population  = 

∑ Pop.age=100+
age=65

∑ Pop.age<65
age=15



The Impact of Migration on Population Ageing in Asia 1990-2020    • 5

remaining life expectancy. The most common definition of the prospective old-age 
threshold is the age when the remaining life expectancy equals 15 years (Sanderson/
Scherbov 2013). Since life expectancy varies over time and across populations, this 
threshold is dynamic. However, Sanderson and Scherbov (2020) have shown that 
death rates, as an indicator of health, are approximately constant at this threshold 
across populations over time and space. The prospective old-age thresholds by 
country for the years 1990 and 2020 are reported in Appendix Table A1. The old-
age population is defined for each country and year as the population at age=POAT 
and older, while the working-age population includes the population aged 15 and 
over but younger than the prospective old-age threshold. The prospective old-age 
dependency ratio (POADR) is then the ratio between the two age groups according 
to the prospective definition:

2.3 Decomposition analysis of changes in the prospective old-age 
dependency ratio

To assess the drivers of age-structure changes in Asian countries, I conduct a 
decomposition analysis of changes in the prospective old-age dependency ratio 
over the period 1990-2020. The end of the period is set at the beginning of year 2020 
to avoid any effects of excess mortality during the Covid-19 pandemic starting in 
2020 (World Health Organization (WHO) 2023). Life expectancy for 2020 is obtained 
from the 2019 life table. Similar to decomposition analyses based on chronological 
age (de Beer et al. 2011; Ghio et al. 2022; Kashnitsky et al. 2017), cohort turnover, 
deaths and migration effects are taken into account. First, the cohort turnover effect 
depends on the number of individuals entering working-age and the number of 
individuals leaving working-age and entering old-age. Thus, the cohort turnover 
effect is an indicator of the age structure, which in turn is the long-term result of 
fertility, mortality and migration. Second, the deaths effect is defined by the number 
of deaths in working-age and old-age, respectively. Third, the migration effect is 
based on the number of net migrants of working-age and in old-age, respectively. 

When using prospective age, the definition of the relevant age groups shifts 
with changing life expectancy. Hence, I extend the decomposition method by 
adding an effect which accounts for the additional cohort turnover due to changes 
in the prospective old-age threshold. In case of life expectancy improvements, 
the prospective old-age threshold – the age at which the average remaining life 
expectancy equals 15 years – increases. In such instances, the effect of changes in life 
expectancy (LE) is determined by the additional number of individuals that “remain” 
in the working-age population and do not enter old-age due to the increase in 
the prospective old-age threshold. When life expectancy decreases, as observed at 
least temporarily in some countries, the effect is defined by the additional number 
of individuals that leave working-age and enter old-age due to the reduction in the 

Prospective old-age dependency ratio (POADR) = prosp. old-age population
prosp. working-age population  = 

∑ Pop.age=100+
age=POAT

∑ Pop.age<POAT
age=15



•    Markus Dörflinger6

prospective old-age threshold. It should be noted that changes in life expectancy 
imply changes in mortality rates, which, of course, in turn impact the effect of 
deaths. In this paper, however, the effect of changes in life expectancy refers only to 
the effect of changes in the average remaining life expectancy and thus changes in 
the prospective old-age threshold.

Equation 1 describes the changes in the prospective old-age population (POA; 
numerator of the POADR) from the beginning of year t to the beginning of year t+1 
considering the prospective old-age thresholds in the two consecutive years (POAT(t) 
and POAT(t+1)). The changes in the POA are attributed to the four components 
described above:

where CTPOA(t) is the cohort turnover in old-age, DPOA(t) is the number of deaths in 
old-age, MPOA(t) is the net migration in old-age and CTLEchange(t) is the additional 
cohort turnover into old-age due to changes in life expectancy. 

Similarly, equation 2) describes the changes in the prospective working-age 
population (PWA; denominator of the POADR) from the beginning of year t to the 
beginning of year t+1:

where CTPWA(t) is the cohort turnover in working-age minus the cohort turnover 
in old-age, DPWA(t) is the number of deaths in working-age, MPWA(t) is the net 
migration in working-age and CTLEchange(t) is the additional cohort turnover in 
old-age due to changes in life expectancy. 

I assume that all migration takes place at the end of the year, which is a reasonable 
assumption if data are available by single year (UN DESA 2022a). Accordingly, 
migrants do not contribute to either cohort turnover or deaths occurring in the year 
of migration. Cohort turnover due to changes in life expectancy at the end of year 
CTLEchange(t) is positive if the prospective old-age threshold increases and negative 
if it decreases. If it remains stable, the effect size equals 0. The following example 
illustrates how changes in the prospective old-age threshold impact the equations 
above. We assume an increase of the prospective old-age threshold from 67 years 
(2000) to 68 years (2001). Hence, CTLEchange(2000) comprises all individuals aged 
67 at the end of year 2000 – let us assume 100 individuals – since they would have 
entered old-age at the beginning of year 2001 if the prospective old-age threshold 

1)

2)

1) ∆POA(t, t+1) = POA(t+1) − POA(t) = 

2) ∑ Pop.(t+1)age=100+
age=POAT(t+1)  − ∑ Pop.(t) =age=100+

age=POAT(t)

3)   CTPOA(t) − DPOA(t) + MPOA(t) + CTLEchange(t)

  

2) ∆PWA(t, t+1) = PWA(t+1) − PWA(t) = 

3)                                  ∑ Pop.(t+1)age<POAT(t+1)
age=15  − ∑ Pop.(t) =age<POAT(t)

age=15

                                CTPWA(t) − DPWA(t) + MPWA(t) − CTLEchange(t) 

  



The Impact of Migration on Population Ageing in Asia 1990-2020    • 7

remained stable at age 67 but did not because it increased to age 68. These 100 
individuals “remain” in the working-age population also in year 2001. Therefore, with  
CTLEchange(2000) equal to -100, the working-age population in year 2001 is 100 
individuals larger and the old-age population is 100 individuals smaller than it would 
have been without any change in life expectancy. 

Reliable data on international migration are lacking in many countries (Willekens 
et al. 2016). This is particularly the case when data on migration by age are needed, 
as is the case in this analysis. Therefore, I use the residual method to estimate net 
migration (Siegel/Hamilton 1952). Annual net migration by age group at the end of 
the year t is derived from equations (1) and (2) above:5

Using these definitions of the annual changes in the size of the old-age and 
working-age populations, changes in the prospective old-age dependency ratio are 
decomposed using a two-step decomposition of a rate (Kashnitsky et al. 2017). First, 
changes in the prospective old-age dependency ratio are decomposed into changes 
in the size of the numerator (old-age population) and the denominator (working-
age population). Second, each of these changes is further decomposed into the 
effects of the four demographic components introduced above (cohort turnover, 
deaths, migration, changes in life expectancy). In contrast to Kashnitsky et al. (2017), 
who focused only on changes in the working-age population, I decompose both the 
numerator and denominator effects, considering the effects of the demographic 
components on changes in the working-age population and changes in the old-age 
population.

In the first step, I make use of the formula for the decomposition of rates as a 
function of two factors (Das Gupta 1991): 

The first of the two terms on the right-hand side of equation 5) represents the 
α-effect, and the second represents the β-effect. The α-effect describes how changes 
in the size of the prospective old-age population (numerator of the ratio) affect the 
POADR. The β-effect describes how changes in the size of the prospective working-

5 The annual sum of the derived net migration by age group corresponds to the total annual net 
migration provided by the UN. 

3) MPOA(t) = POA(t+1) − POA(t) − CTPOA(t) +  DPOA(t) − CTLEchange(t) 

  

3)

4) MPWA(t) = PWA(t+1) − PWA(t) − CTPWA(t) +  DPWA(t) + CTLEchange(t) 
 

  

4)

5) POADR(t+1) − POADR(t) = α β
6) [1

2 ∗ (POA(t+1) − POA(t)) ∗ ( 1
PWA(t+1) + 1

PWA(t))] +   

                                                        [1
2 ∗ (POA(t+1) + POA(t)) ∗ ( 1

PWA(t+1) − 1
PWA(t))]

5)



•    Markus Dörflinger8

age population (denominator of the ratio) affect the POADR (Das Gupta 1993). The 
average α-effect across all 51 countries was -5.3 percentage points with a standard 
deviation of 3.0 percentage points. The average β-effect was +5.4 percentage points 
with a standard deviation of 2.3 percentage points. 

In the second step of the decomposition, I further decompose the two effects 
from equation 5) into the effects of cohort turnover, deaths, migration and changes 
in life expectancy. First, I replace POA(t+1) − POA(t) in the formula for the α-effect 
with equation 1) (see Appendix Equation 1 for the step-by-step substitution):

The four right-hand terms of equation 6) denote the effects of cohort turnover in 
old-age, deaths in old-age, net migration in old-age and changes in life expectancy 
on the numerator of the prospective old-age dependency ratio. A similar 
transformation is done with the formula for the β-effect, which is rearranged to 
replace PWA(t+1) − PWA(t) with equation 2) (see Appendix Equation 2 for the step-
by-step substitution). The four right-hand terms of equation 7) denote the effects 
of cohort turnover as the difference between the number of individuals entering 
and leaving the working-age population, deaths in the working-age population, net 
migration in the working-age population and changes in life expectancy on the 
denominator of the prospective old-age dependency ratio:

Finally, the effects by component (CT, D, M, CTLEchange) are obtained by summing 
the α- and β-effects by component from equations 6) and 7). The decomposition 
analysis is performed for the entire period from 1990 to 2020 and separately by 
decade to illustrate changes over time.

6)6) α [1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ CTPOA(t)] + 

      [− 1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ DPOA(t)] + 

      [1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ MPOA(t)] + 

      [1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ CTLEchange(t)] 

7) β [− 1
2 ∗ POA(t+1)+POA(t)

PWA(t+1)∗PWA(t) ∗ CTPWA(t)] + 

       [1
2 ∗ POA(t+1)+POA(t)

PWA(t+1)∗PWA(t) ∗ DPWA(t)] + 

       [− 1
2 ∗ POA(t+1)+POA(t)

PWA(t+1)∗PWA(t) ∗ MPWA(t)] + 

       [1
2 ∗ POA(t+1)+POA(t)

PWA(t+1)∗PWA(t) ∗ CTLEchange(t)] 

  

7)



The Impact of Migration on Population Ageing in Asia 1990-2020    • 9

3 Results

3.1 Population ageing 1990-2020

The level of population ageing varies substantially across Asian countries. For instance, 
the countries of the Gulf Cooperation Council (GCC) – Bahrain, Kuwait, Oman, Qatar, 
Saudi Arabia and the United Arab Emirates – have an extremely young age structure, 
while Japan has one of the oldest populations in the world (Sanderson/Scherbov 
2008). However, in countries with high life expectancy, such as Japan, population 
ageing seems less dramatic when using prospective measures. Table 1 shows the 
prospective old-age threshold (POAT), the prospective old-age dependency ratio 
(POADR) and the chronological old-age dependency ratio (OADR) using age 65 
as the old-age threshold. The prospective old-age threshold – the age at which 
the average remaining life expectancy equals 15 years – ranges from age 62.1 in 
Afghanistan to age 74.7 in Hong Kong. In 16 of the 51 countries, the prospective old-
age threshold is lower than the classical old-age threshold at age 65. Consequently, 
the POADR is higher than the OADR. However, in 35 countries, the prospective old-
age threshold is higher than 65, leading to a lower prospective old-age dependency 
ratio. While the differences in the level of population ageing between countries 
are less pronounced when applying the prospective old-age threshold, they remain 
remarkable. On one hand, the GCC countries exhibit the world’s lowest prospective 
old-age dependency ratios, ranging from 0.7 percent (Qatar) to 2.9 percent (Kuwait). 
However, low or relatively low prospective old-age dependency ratios are not found 
solely in Western Asia, but also in other subregions, e.g. in Macao (5.5 percent), the 
Maldives (4.2 percent), Singapore (5.6 percent) and Afghanistan (6.1 percent). On the 
other hand, Japan and Georgia reach high values of 23.0 percent and 22.2 percent 
respectively, ranking them among the world’s most aged countries (UN DESA 2022b, 
own calculations). Following these two exceptional countries, other countries with 
higher prospective old-age dependency ratios are spread over all subregions, such 
as Armenia (15.6 percent), Kazakhstan (14.1 percent), Myanmar (12.2 percent), North 
Korea (13.4 percent) and Sri Lanka (13.2 percent).

The age structure of Asian countries has also evolved quite differently in recent 
decades. In 27 of the 51 countries, population ageing – measured by an increase in 
the prospective old-age dependency ratio – has advanced between 1990 and 2020. 
Japan stands out with an increase of more than eleven percentage points (Fig. 1), 
followed by Georgia (+7.0 percentage points), North Korea (+6.1 percentage points), 
Armenia (+6.1 percentage points) and Sri Lanka (+5.0 percentage points). Globally, 
only about 20 countries – mainly in Europe and the Caribbean – have experienced 
comparable increases in the prospective old-age dependency ratio during this period 
(UN DESA 2022b, own calculations), underlining the notable speed of population 
ageing in some Asian countries. In 24 of 51 Asian countries however, the prospective 
old-age dependency ratio decreased from 1990 to 2020, e.g. in Afghanistan (-4.1 
percentage points), Iraq (-4.4 percentage points) and Mongolia (-4.9 percentage 
points). Similar declines are seen in many other countries, particularly in sub-
Saharan Africa. A decreasing prospective old-age dependency ratio is not surprising 



•    Markus Dörflinger10

Tab. 1: Prospective old-age threshold, prospective old-age dependency ratio 
(POADR) and chronological old-age dependency ratio (OADR) (2020)

Prospective old-age POADR OADR
threshold

Central Asia
Kazakhstan 63.8 14.1 12.3
Kyrgyzstan 63.9 7.9 6.9
Tajikistan 63.8 6.2 5.3
Turkmenistan 63.3 8.9 7.2
Uzbekistan 62.9 9.4 7.3

Eastern Asia
China 68.6 11.7 17.7
Hong Kong 74.7 10.2 26.5
Japan 74.1 23.0 50.3
Macao 73.9 5.5 15.3
Mongolia 65.2 6.4 6.6
North Korea 66.9 13.4 15.7
South Korea 72.9 10.2 21.3
Taiwan 71.7 10.1 20.5

South-Eastern Asia
Brunei Darussalam 66.2 6.3 7.4
Cambodia 65.4 7.5 7.9
Indonesia 63.8 11.2 9.9
Lao 63.0 8.2 6.6
Malaysia 66.8 8.1 9.9
Myanmar 62.5 12.2 9.3
Philippines 64.3 8.7 8.1
Singapore 73.3 5.6 16.9
Thailand 73.8 7.7 19.2
Timor-Leste 63.0 10.8 9.3
Viet Nam 68.0 8.5 12.0

Southern Asia
Afghanistan 62.1 6.1 4.5
Bangladesh 65.8 7.6 8.3
Bhutan 64.0 9.4 8.5
India 65.9 8.9 9.8
Iran 67.4 7.8 10.1
Maldives 68.8 4.2 5.8
Nepal 63.1 11.4 9.5
Pakistan 63.1 8.7 7.1
Sri Lanka 67.0 13.2 16.2



The Impact of Migration on Population Ageing in Asia 1990-2020    • 11

at an earlier stage of population ageing, even though the chronological old-age 
dependency ratio might already be increasing at the same time (Sanderson/Scherbov 
2008). 

An examination of the changes in the prospective old-age dependency ratio by 
decade (Fig. 1) indicates that in most of the Asian countries with an overall decline, 
the negative trend has either diminished over time, e.g. in India, or even shifted 
towards an increase in the 2010s, e.g. in Iran. A distinct development emerges in 
Kazakhstan and Uzbekistan. In these Central Asian countries, the prospective old-
age dependency ratio rose sharply in the 1990s and declined in the 2000s before 
increasing again in the last decade. Also in Armenia and Georgia (Western Asia), 
remarkable increases in the prospective old-age dependency ratio have been 
observed during the 1990s. As the following decomposition analysis reveals, this 
development is partly due to migration.

Tab. 1: Continuation

Prospective old-age POADR OADR
threshold

Western Asia
Armenia 66.4 15.6 18.2
Azerbaijan 65.1 9.1 9.2
Bahrain 69.1 2.3 4.0
Cyprus 70.6 12.1 20.0
Georgia 65.1 22.2 22.3
Iraq 64.2 6.3 5.9
Israel 72.4 9.7 19.6
Jordan 66.7 4.7 5.6
Kuwait 68.7 2.9 5.0
Lebanon 68.9 10.2 14.6
Oman 67.7 2.8 3.7
Palestine 66.2 5.3 6.0
Qatar 69.7 0.7 1.4
Saudi Arabia 68.0 2.3 3.2
Syria 65.0 7.5 7.5
Turkey 68.2 8.6 11.9
United Arab Emirates 69.2 1.3 1.9
Yemen 62.8 5.9 4.8

Notes: Prospective old-age threshold: Age at which the average remaining life expectancy 
equals 15 years.
Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger12

3.2 Drivers of population ageing 1990-2020

Turning towards the drivers of the age-structure changes illustrated above,Figure 2 
depicts the total change in the prospective old-age dependency ratio (x-axis) and 
the change due to cohort turnover, deaths, changes in life expectancy and migration 
(y-axis) over the period 1990-2020. Detailed values by country are reported in 
Appendix Table A2. In general, cohort turnover and deaths had a larger impact 
on the prospective old-age dependency ratio than changes in life expectancy and 
migration. The effect of cohort turnover, driven mainly by growing cohorts entering 
old-age, is positive in all countries. The largest effects are found in Georgia (+38.7 
percentage points) and Japan (+44.5 percentage points), indicating an age structure 
increasingly dominated by older cohorts. Conversely, the effect of cohort turnover 
is smaller in countries with younger age structures and larger cohorts entering 
working-age. This is the case in countries with still relatively high fertility rates, e.g. 
Palestine (+11.9 percentage points) and Yemen (+11.6 percentage points), as well as 
in the GGC countries where the age structure is skewed towards working-age due to 
labour migration. The smallest effect is observed in the United Arab Emirates (+3.3 
percentage points).

While cohort turnover accelerated population ageing, deaths consistently yielded 
a decelerating impact on population ageing across all countries. This is because 
mortality happens predominantly at older ages and less at working-age, consequently 
reducing the prospective old-age dependency ratio. As with the cohort turnover 
effect, this effect is most pronounced in Georgia (-33.7 percentage points) and Japan 
(-26.2 percentage points) and least pronounced in the GCC countries, such as the 
United Arab Emirates (-2.2 percentage points). These differences arise because the 

Fig. 1: Change in the prospective old-age dependency ratio (1990-2020) (in 
percentage points)

1990−2020

−5.0

−2.5

0.0

2.5

5.0

7.5

10.0

1990−2000

−2.5

0.0

2.5

5.0

2000−2010

−2.5

0.0

2.5

5.0

2010−2020

−2.5

0.0

2.5

5.0

Map Source: Minnesota Population Center 2020.
Source: UN DESA 2022b, own calculations.



The Impact of Migration on Population Ageing in Asia 1990-2020    • 13

number of deaths by age group – and consequently the effect of deaths on the 
prospective old-age dependency ratio – is not only determined by the age-specific 
death rates, but also depends heavily on the age structure (see Liang et al. 2023). 
This can be illustrated by applying the age structure of the United Arab Emirates in 
1990, which was the youngest in that year, to all countries. The standardized number 
of deaths in all countries – except the United Arab Emirates of course – is lower than 
the actual number of deaths observed in 1990. In Qatar and Kuwait, only about 
5 percent of deaths in 1990 can be attributed to the slightly older age structure of 
these countries compared with the United Arab Emirates 1990 standard (results not 
shown here). In Japan, however, this percentage reaches 81 percent, indicating the 
large impact of Japan’s old age structure on the number of deaths. These examples 
underline that the effect of deaths on the prospective old-age dependency ratio 
reflects to a large part the country-specific age distribution. 

The effect of changes in life expectancy on the prospective old-age dependency 
ratio is determined by changes in the prospective old-age threshold. With the 
exception of Uzbekistan, where the prospective old-age threshold in 2020 was at 
about the same level as in 1990, it increased in all Asian countries between 1990 and 
2020 (Appendix Table A1). Consequently, this led to a reduction in the age groups 
defined as the old-age population and resulted in a negative effect of changes in life 
expectancy on the prospective old-age dependency ratio (Fig. 2). The effect is most 
pronounced in countries with large improvements in life expectancy, particularly 

Fig. 2: Total change in the prospective old-age dependency ratio and change 
by component (1990-2020) (in percentage points)

Cohort turnover

−30

−20

−10

0

10

20

30

40

−5 0 5 10
Total

Co
ho

rt
 tu

rn
ov

er

Deaths

−30

−20

−10

0

10

20

30

40

−5 0 5 10
Total

D
ea

th
s

Changes in
life expectancy

−30

−20

−10

0

10

20

30

40

−5 0 5 10
Total

Ch
an

ge
s i

n 
lif

e 
ex

pe
ct

an
cy

Migration

−30

−20

−10

0

10

20

30

40

−5 0 5 10
Total

M
ig

ra
tio

n

Central Asia
Eastern Asia
South−Eastern Asia
Southern Asia
Western Asia

Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger14

in Eastern Asia (e.g. Hong Kong and South Korea). In South Korea, for instance, the 
prospective old-age threshold increased over the three decades from age 65.3 to 
age 72.9, which in turn decreased the prospective old-age dependency ratio by seven 
percentage points (Appendix Table A2). Conversely, the effect is smaller in countries 
with minor differences in the prospective old-age threshold, e.g. in the Central 
Asian countries Turkmenistan and Uzbekistan. In Qatar and other GCC countries in 
Western Asia, the impact also remains modest despite noticeable improvements 
in life expectancy. This is again due to their particularly young population. Since 
the effect of changes in life expectancy also depends on the number of individuals 
potentially affected by a shift in the old-age threshold, the small cohort sizes at 
these ages reduce the effect of increasing old-age thresholds on the prospective 
old-age dependency ratio.

Turning towards the impact of migration, both a decelerating and an accelerating 
effect on population ageing can be found in Asian countries (Fig. 2). A decelerating 
effect is due to positive net migration at working-age and/or negative net migration 
at old-age, while an accelerating effect is based on the opposite patterns. Despite 
the migration effect being smaller than the effect of deaths and cohort turnover 
in most countries, migration had a notable impact on population ageing in some 
countries over the last three decades. Figure 3 shows how migration has contributed 
to changes in the prospective old-age dependency ratio from 1990 to 2020. On 
one hand, large negative effects are observed in high-immigration countries, e.g. 
in Bahrain (-4.5 percentage points), Macao (-3.6 percentage points), Oman (-4.5 
percentage points) and Singapore (-4.4 percentage points). In these countries, the 
prospective old-age dependency ratio would have increased without migration, 
illustrating that migration has not only slowed down but halted the process of 
population ageing. This migration effect is driven largely by high net migration 
at working-age (results not shown here), which aligns with the higher intensity of 
migration among younger adults. On the other hand, the largest positive effects 
of migration on the prospective old-age dependency ratio are found in Armenia 
(+4.8 percentage points), Georgia (+3.7 percentage points) and Timor-Leste (+7.2 
percentage points). All three countries are characterized by high negative net 
migration rates (Appendix Table A6). While in Timor-Leste this migration effect 
was largely offset by other factors, particularly substantial improvements in life 
expectancy (Appendix Table A2), migration remained a pivotal driver of rapid 
population ageing in Armenia and Georgia. 

As the above examples illustrate, differences in population ageing, at least 
between some high-immigration and high-emigration countries, are substantially 
driven by migration. In most Asian countries, however, the impact of migration on 
population ageing was relatively small. For instance, in Japan, which experienced 
the largest increase in the prospective old-age dependency ratio between 1990 
and 2020, the effect of slightly positive net migration rates on the rapidly advancing 
ageing process was marginal. But even in countries with remarkable emigration flows, 
such as Bangladesh, India and the Philippines (Abel et al. 2019), the migration effect 
is small. This may be attributed to temporary migration, which is a common pattern 
in Asia (Hugo 2012). This analysis captures the effect of net migration over the entire 



The Impact of Migration on Population Ageing in Asia 1990-2020    • 15

period under examination. For instance, a person leaving the country at working-
age contributes to a positive effect on the old-age dependency ratio in the year of 
emigration. If this person returns before reaching old-age, the effect in the year of 
return is negative. The opposite effects are observed in the destination country. In 
general, if immigration and emigration occur in different years but within the same 
age group, the effects of immigration and emigration balance out over time. Since 
this analysis relies on population changes between two consecutive years, it does 
not capture short-term, seasonal mobility within the same year. However, temporary 
migrants from South-Eastern Asia typically stay in destination countries for several 
years (Bossavie et al. 2021; Valenta et al. 2020). Nevertheless, they may remain 
unregistered and migration flows may not be fully captured in the UN population 
data. In general, data on net migration by age group, derived as a residual from 
population and vital statistics, depend on the quality of the underlying data. In 
this analysis, negative migration effects on the prospective old-age dependency 
ratio emerge in China and India, suggesting positive net migration at working-age 
and/or negative net migration at old-age. However, this seems unlikely given the 
slightly negative net migration rates in both countries, which are mainly driven by 
high emigration in young adult ages (Shen et al. 2024). Rather, these effects may be 
related to inaccuracies in the underlying data on population by age and age-specific 
deaths. For instance, death registration rates in both countries are lower than in 
most developed countries (Karlinsky 2024). An under-reporting of deaths in old-

Fig. 3: Change in the prospective old-age dependency ratio due to migration 
(1990-2020) (in percentage points)

−5.0

−2.5

0.0

2.5

5.0

7.5

10.0

Map Source: Minnesota Population Center 2020. 
Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger16

age may lead to lower net migration in this age group, also impacting the overall 
migration effect on population ageing.

The impact of cohort turnover, deaths, changes in life expectancy and migration 
on the prospective old-age dependency ratio varies over time. The results of the 
decomposition by decade are reported in the Appendix (Table A3-5).6 In all countries, 
cohort turnover had a consistently positive effect and deaths a consistently negative 
effect on the POADR. However, the impact of cohort turnover and deaths has 
increased in most countries towards the last decade, indicating an increasing shift 
towards older age structures. This is particularly true for rapidly ageing countries 
such as Japan, while in countries with a younger age structure, such as the United 
Arab Emirates, the impact of these components remains stable or even decreases 
slightly over the decades.

While the effect of changes in life expectancy over the entire period 1990-2020 
is negative in all countries, reflecting improvements in life expectancy, the analysis 
by decade reveals that this trend is not uniformly linear in all countries. In some 
countries, temporary decreases in life expectancy are observed, leading to positive 
effects of changes in life expectancy on the POADR. The reversal of the effect of 
changes in life expectancy in many Western and Central Asian countries in the 
1990s is particularly striking. The post-Soviet countries were strongly affected by the 
political, economic and social upheavals of the transformation phase. In most former 
Soviet republics, life expectancy declined in the 1990s before gradually recovering 
(UN DESA 2022b; Duthé et al. 2017; Grigoriev et al. 2014). This accelerated – at least 
temporarily – population ageing in the 1990s. In Kazakhstan, about 51 percent (+2.4 
of +4.6 percentage points) of the increase in the prospective old-age dependency 
ratio between 1990 and 2000 can be attributed to declining life expectancy. 
Similar effects can be observed in North Korea (+2.6 percentage points), where life 
expectancy fell sharply in the 1990s due to famine and political mismanagement 
(Spoorenberg/Schwekendiek 2012). More recently, the Syrian civil war after 2011 
led to a temporary decline in life expectancy. However, life expectancy and the 
prospective old-age threshold recovered rapidly, so that the effect of these changes 
on the POADR in the 2010s remains small (+0.1 percentage points). 

There is also considerable variation over time in the impact of migration on 
age-structure changes. International migration flows within, from and to Asia are 
driven by various cultural, demographic, economic, educational, historical as well 
as conflict- and network-related factors (Abel et al. 2019). In addition, migration 
policies and legal frameworks play an important role in shaping migration (de 
Haas et al. 2019). Because all of these factors are subject to change, net migration 
rates vary over time in many countries (Appendix Table A6). Figure 4 shows net 
migration rates from 1990 to 2019 for three selected countries: Armenia, Syria and 

6 The sum of each of the four effects over the decades does not necessarily equal the corresponding 
effect calculated for the entire period 1990-2020, since the effect size depends on the changes in the 
size of the population by age group over the period under consideration, which in most cases follow 
a non-linear pattern.



The Impact of Migration on Population Ageing in Asia 1990-2020    • 17

the United Arab Emirates. Net migration rates in the United Arab Emirates reached a 
pronounced peak around 2005 to 2009, followed by a significant decline associated 
with the lower demand for labour migrants following the global economic crisis (de 
Bel-Air 2018). In the 2010s, net migration in the United Arab Emirates even turned 
slightly negative. As a result, the overall migration effect from 1990 to 2020 is lower 
in the United Arab Emirates than in other GCC countries. Migration and its impact 
on population ageing can also be temporarily impacted by armed conflict and 
displacement. In Syria, net migration rates peaked in 2006 and 2007, related to large 
refugee movements from neighbouring Iraq (Lischer 2008), and fell to negative rates 
with the onset of the Syrian civil war in 2011, which led to large refugee emigration 
(McAuliffe/Triandafyllidou 2021). These events had a noticeable impact on the age 
structure. About 56 percent (+1.1 of +2.0 percentage points) of the increase in the 
prospective old-age dependency ratio in Syria during the 2010s can be attributed 
to migration. Similar dynamics can be observed in other Asian countries affected by 
armed conflict, such as Afghanistan, Iraq and Timor-Leste. The downfall of the Soviet 
Union also had a remarkable impact on net migration in the former Soviet republics. 
In Armenia, for instance, emigration rose remarkably after gaining independence 
in 1991, contributing to the large increase in the prospective old-age dependency 
ratio in the 1990s. The migration effect accounted for about 35 percent (+2.8 of 
+6.3 percentage points) of the increase between 1990 and 2000. These examples 
illustrate the impact of temporary increases or declines in net migration on changes 
in the age structure of Asian countries.

Fig. 4: Net migration rate for selected countries (1990-2019) (per 1,000 
population)

Armenia

Syria

United Arab Emirates

−100

0

100

200

1990 1995 2000 2005 2010 2015 2020
Year

N
et

 m
ig

ra
tio

n 
ra

te
 (p

er
 1

,0
00

 p
op

ul
at

io
n)

Source: UN DESA 2022b.



•    Markus Dörflinger18

4 Conclusion and discussion

Population ageing in Asia has evolved differently over recent decades, with some 
countries ageing remarkably and others maintaining a very young age structure. In 
this paper, I assess the impact of international migration on age-structure changes 
in 51 Asian countries from 1990 to 2020. To take into account varying health and life 
expectancy, I propose a new approach to incorporate a dynamic prospective old-
age threshold into a decomposition analysis of changes in the old-age dependency 
ratio. The results reveal a smaller effect of migration on population ageing compared 
to the effects of cohort turnover and deaths. However, extraordinary migration 
effects are observed in some countries with either very negative or very positive 
net migration. As migration largely occurs at younger ages, high immigration had 
a decelerating effect on population ageing, and high emigration an accelerating 
effect. Hence, the large differences in the speed and level of population ageing 
across Asian countries can be partly attributed to migration. 

The findings reflect both the evolution and variation of migration policies 
as well as geopolitical and economic changes that have influenced migration 
patterns since 1990. Economic growth and policies promoting temporary labour 
migration in the countries of the Gulf Cooperation Council (Aarthi/Sahu 2021) led 
to high immigration flows at working-age, which slowed down or even halted the 
process of population ageing. Other economically strong countries, such as South 
Korea, remained relatively closed to immigration due to predominantly restrictive 
migration policies (de Haas et al. 2018), so that almost no such decelerating effect 
of migration on population ageing was observed. Yet, countries like Japan and 
South Korea are increasingly making efforts to open up to more immigration, also 
driven by the demographic challenges posed by an ageing population (de Haas 
et al. 2018; Oishi 2021). Finally, the dissolution of the Soviet Union accelerated 
emigration from former Soviet republics in the 1990s, which contributed to faster-
ageing populations. This effect is most pronounced in Armenia and Georgia, 
despite policy efforts to constrain emigration (Makaryan/Chobanyan 2014; Reslow 
2017). However, both countries have recently experienced higher immigration flows, 
at least temporarily halting the longstanding trend of negative net migration. In 
Georgia, this is attributed to a rise in migration from Russia following the invasion in 
Ukraine in 2022 (National Statistics Office of Georgia 2024). In Armenia, it is driven by 
refugees from Azerbaijan (Nagorno-Karabakh) following the Azerbaijani offensive 
in Artsakh in 2023 (Statistical Committee Republic of Armenia 2024).

This study introduces an approach to incorporate the prospective age concept 
into decomposition analyses of age-structure changes. Prospective age takes into 
account differences in life expectancy – as a proxy for additional characteristics 
beyond chronological age – thus providing a more comprehensive picture of 
population ageing. Although the impact of changes in life expectancy on the old-age 
dependency ratio remains modest compared to the effects of cohort turnover and 
deaths, the application of prospective age in the decomposition of age-structure 
changes increases comparability across different contexts. Further studies using the 
approach presented in this paper may yield valuable insights into the drivers behind 



The Impact of Migration on Population Ageing in Asia 1990-2020    • 19

age-structure changes in other world regions and countries applying the prospective 
view of population ageing. Prospective age is calculable for all countries, making it 
a very useful measure for such comparative analyses. It should be noted, however, 
that country-level analyses do not consider health differences across population 
subgroups. For example, it is well known that education is strongly correlated with 
higher (healthy) life expectancy (Weber/Loichinger 2022). Moreover, the actual 
expected lifetime and prospective old-age threshold may be underestimated 
based on period life expectancy, making population ageing appear slightly more 
pronounced compared to when using cohort life expectancy (Kjærgaard/Canudas-
Romo 2017).

The analyses in this study are based on estimates from the UN World 
Population Prospects 2022, which provide a unique source of population data for 
all Asian countries. However, the quality of the UN data depends on the national 
data available and are subject to uncertainties, particularly in countries without 
longstanding high-quality vital registration systems (Liu/Raftery 2020). In Asia, death 
registration is incomplete in most countries (Karlinsky 2024). In addition, temporary 
and undocumented migration flows may not be captured in official statistics 
(Raymer et al. 2022), which might also impact population counts. In general, data 
on international migration, particularly by age, are scarce (Willekens et al. 2016). In 
this study, net migration by age group is derived from population and death counts 
(Siegel/Hamilton 1952). Since this method does not require data on migrant stocks 
or flows, it can be applied even in cases of poor data availability (UN DESA 2022a). 
However, the derived net migration is still subject to uncertainty, as it depends on 
the quality of vital statistics and population counts. Therefore, the results should be 
interpreted with caution, acknowledging the uncertainties in the underlying data.

Another limitation arises from the decomposition method used in this study, 
as it captures only the direct effect of net migration on the age structure, without 
accounting for potential indirect or long-term demographic effects. This is because 
migrants contribute to the migration effect on population ageing only in the year 
of migration. In subsequent years, they become part of the total population of the 
country of residence. As a result, they may potentially contribute to cohort turnover 
and deaths (and again to the migration effect). In addition to these effects, migrants 
can also contribute to age-structure changes through fertility in the long term 
(Goldstein 2009). Given that the fertility of temporary labour migrants, who make 
up a large share of migrants in Asia, is usually lower in the destination countries, the 
long-term effect of immigration through births can be assumed to be lower in many 
Asian countries than in the European context (see Fihel et al. 2018; Murphy 2017). Yet, 
in some Asian countries, such as Armenia and Georgia, the emigration of younger 
adults has had a remarkable impact on the number of births (Weinar 2014), affecting 
the age structure in the long run.

While population ageing seems less dramatic through the lens of prospective 
age, it remains a key issue for development in Asia. However, as this study underlines, 
there is substantial demographic heterogeneity across countries. In addition, the 
level and speed of population ageing also varies remarkably within Asian countries 
(Gietel-Basten et al. 2016). These differences may be related to internal and 



•    Markus Dörflinger20

international migration. In many Asian countries, rural-urban migration is a notable 
phenomenon (Bell et al. 2020). Moreover, international migration patterns also vary 
between urban and rural areas (Hugo 2016). Future studies could further explore 
the impact of both international and internal migration on regional age-structure 
changes, enhancing our understanding of the relationship between migration and 
population ageing in Asian countries.

Acknowledgements
Part of the research was developed in the Young Scientists Summer Program at 
the International Institute for Applied Systems Analysis, Laxenburg (Austria) with 
financial support from the German National Member Organization.

References 

Aarthi, S. V.; Sahu, Mrutuyanjaya 2021: Migration Policy in the Gulf Cooperation Council (GCC) 
States: A Critical Analysis. In: Contemporary Review of the Middle East 8,4: 410-434. 
https://doi.org/10.1177/23477989211028748

Abel, Guy J.; Raymer, James; Guan, Qing 2019: Driving factors of Asian international migration 
flows. In: Asian Population Studies 15,3: 243-265. 
https://doi.org/10.1080/17441730.2019.1664526

Alho, Juha M. 2008: Migration, fertility, and aging in stable populations. In: Demography 45,3: 
641-650. https://doi.org/10.1353/dem.0.0021

Ali, Abdulaziz; Cochrane, Logan 2024: Residency and citizenship in the Gulf: recent policy 
changes and future implications for the region. In: Comparative Migration Studies 12,16: 
1-17. https://doi.org/10.1186/s40878-024-00376-1

Alvarez, Jesús-Adrián; Vaupel, James W. 2023: Mortality as a Function of Survival. In: 
Demography 60,1: 327-342. https://doi.org/10.1215/00703370-10429097

Balachandran, Arun et al. 2020: Comparison of Population Aging in Europe and Asia Using a 
Time-Consistent and Comparative Aging Measure. In: Journal of Aging and Health 32,5-6: 
340-351. https://doi.org/10.1177/0898264318824180

Bell, Martin et al. 2020: Internal Migration in the Countries of Asia. A Cross-national 
Comparison. Cham: Springer International Publishing. 
https://doi.org/10.1007/978-3-030-44010-7

Bijak, Jakub; Kupiszewska, Dorota; Kupiszewski, Marek 2008: Replacement Migration Revisited: 
Simulations of the Effects of Selected Population and Labor Market Strategies for the Aging 
Europe, 2002-2052. In: Population Research and Policy Review 27,3: 321-342. 
https://doi.org/10.1007/s11113-007-9065-2

Billari, Francesco C.; Dalla-Zuanna, Gianpiero 2011: Is Replacement Migration Actually Taking 
Place in Low Fertility Countries? In: Genus 67,3: 105-123. 

Blanchet, Didier 1989: Regulating the age structure of a population through migration. In: 
Population 44,1: 23-37. 

Bossavie, Laurent et al. 2021: Temporary Migration for Long-term Investment. Policy Research 
Working Paper 9740. Washington D.C.: World Bank.



The Impact of Migration on Population Ageing in Asia 1990-2020    • 21

Canudas-Romo, Vladimir; Shen, Tianyu; Payne, Collin 2021: The role of reductions in old-age 
mortality in old-age population growth. In: Demographic Research 44: 1073-1084.
https://doi.org/10.4054/DEMRES.2021.44.44

Caselli, Graziella; Vallin, Jacques 1990: Mortality and population ageing. In: European Journal 
of Population 6,1: 1-25. https://doi.org/10.1007/BF01796797

Coale, Ansley J. 1957: How the Age Distribution of a Human Population is Determined. In: 
Cold Spring Harbor Symposia Quantitative Biology 22: 83-89. 
https://doi.org/10.1101/SQB.1957.022.01.010

Coale, Ansley J. 1986: Demographic Effects of Below-Replacement Fertility and Their Social 
Implications. In: Population and Development Review 12: 203-216. 
https://doi.org/10.2307/2807903

Coleman, David 2008: The demographic effects of international migration in Europe. In: 
Oxford Review of Economic Policy 24,3: 452-476. https://doi.org/10.1093/oxrep/grn027

Craveiro, Daniela et al. 2019: Back to replacement migration: A new European perspective 
applying the prospective-age concept. In: Demographic Research 40: 1323-1344.
https://doi.org/10.4054/DemRes.2019.40.45

d’Albis, Hippolyte; Collard, Fabrice 2013: Age groups and the measure of population aging. In: 
Demographic Research 29: 617-640. https://doi.org/10.4054/DemRes.2013.29.23

Das Gupta, Prithwis 1991: Decomposition of the difference between two rates and its 
consistency when more than two populations are involved. In: Mathematical Population 
Studies 3,2: 105-125. https://doi.org/10.1080/08898489109525329

Das Gupta, Prithwis 1993: Standardization and Decomposition of Rates: A User’s Manual. U.S. 
Bureau of the Census, Current Population Reports, Series P23-186. Washington D.C.: U.S. 
Government Printing Office.

de Beer, Joop; van der Erf, Rob; Huisman, Corina 2011: The growth of working age population: 
differences between rural and urban regions across Europe. NEUJOBS Final Draft Working 
Paper D 8.1. The Hague: Netherlands Interdisciplinary Demographic Institute.

de Bel-Air, Françoise 2018: Asian Migration to the Gulf States in the Twenty-first Century. 
In: Chowdhury, Mehdi; Irudaya Rajan, S. (Eds.): South Asian Migration in the Gulf. Cham: 
Springer International Publishing: 7-34. https://doi.org/10.1007/978-3-319-71821-7_2

de Haas, Hein et al. 2019: International Migration: Trends, Determinants, and Policy Effects. 
In: Population and Development Review 45,4: 885-922. https://doi.org/10.1111/padr.12291

de Haas, Hein; Natter, Katharina; Vezzoli, Simona 2018: Growing Restrictiveness or Changing 
Selection? The Nature and Evolution of Migration Policies. In: International Migration 
Review 52,2: 324-367. https://doi.org/10.1111/imre.12288

Demuru, Elena; Egidi, Viviana 2016: Adjusting prospective old-age thresholds by health status: 
empirical findings and implications. A case study of Italy. In: Vienna Yearbook of Population 
Research 14: 131-154. https://doi.org/10.1553/populationyearbook2016s131

Duthé, Géraldine et al. 2017: Adult mortality patterns in the former Soviet Union’s southern 
tier: Armenia and Georgia in comparative perspective. In: Demographic Research 36: 589-
608. https://doi.org/10.4054/DemRes.2017.36.19

Espenshade, Thomas J. 1994: Can Immigration Slow U.S. Population Aging? In: Journal of 
Policy Analysis and Management 13,4: 759-768. https://doi.org/10.2307/3325496

Espenshade, Thomas J.; Bouvier, Leon F.; Arthur, W. Brian 1982: Immigration and the Stable 
Population Model. In: Demography 19,1: 125-133. https://doi.org/10.2307/2061132



•    Markus Dörflinger22

Fernandes, Fernando; Turra, Cássio M.; Rios Neto, Eduardo L.G. 2023: World population aging 
as a function of period demographic conditions. In: Demographic Research 48: 353-372.
https://doi.org/10.4054/DemRes.2023.48.13

Fihel, Agnieszka; Janicka, Anna; Kloc-Nowak, Weronika 2018: The direct and indirect impact 
of international migration on the population ageing process: A formal analysis and its 
application to Poland. In: Demographic Research 38: 1303-1338.
https://doi.org/10.4054/DemRes.2018.38.43

Fihel, Agnieszka; Janicka, Anna; Okólski, Marek 2023: Predicting a Migration Transition 
in Poland and its Implications for Population Ageing. In: Central and Eastern European 
Migration Review 12,1: 265-292. https://doi.org/10.54667/ceemr.2023.15

Ghio, Daniela; Goujon, Anne; Natale, Fabrizio 2022: Assessing the demographic impact of 
migration on the working-age population across European territories. In: Demographic 
Research 46: 261-272. https://doi.org/10.4054/DemRes.2022.46.9

Gietel-Basten, Stuart; Scherbov, Sergei; Sanderson, Warren 2015: Remeasuring Ageing in 
Southeast Asia. In: Asian Population Studies 11,2: 191-210. 
https://doi.org/10.1080/17441730.2015.1052201

Gietel-Basten, Stuart; Scherbov, Sergei; Sanderson, Warren 2016: Towards a reconceptualising 
of population ageing in emerging markets. In: Vienna Yearbook of Population Research 14: 
41-66. https://doi.org/10.1553/populationyearbook2016s041

Global Burden of Disease Collaborative Network 2020: Global age-sex-specific fertility, 
mortality, healthy life expectancy (HALE), and population estimates in 204 countries and 
territories, 1950-2019: a comprehensive demographic analysis for the Global Burden of 
Disease Study 2019. In: Lancet 396,10258: 1160-1203. 
https://doi.org/10.1016/S0140-6736(20)30977-6

Goldstein, Joshua R. 2009: How Populations Age. In: Uhlenberg, Peter (Ed.): International 
Handbook of Population Aging. Dordrecht: Springer Netherlands: 7-18. 
https://doi.org/10.1007/978-1-4020-8356-3_1

Grigoriev, Pavel et al. 2014: The Recent Mortality Decline in Russia: Beginning of the 
Cardiovascular Revolution? In: Population and Development Review 40,1: 107-129.
https://doi.org/10.1111/j.1728-4457.2014.00652.x

Horiuchi, Shiro 1991: Assessing the effects of mortality reduction on population ageing. In: 
Population Bulletin of the United Nations 31/32. New York: United Nations.

Hugo, Graeme 2012: International Labour Migration and Migration Policies in Southeast Asia. 
In: Asian Journal of Social Science 40,4: 392-418. 
https://doi.org/10.1163/15685314-12341250

Hugo, Graeme 2016: Internal and International Migration in East and Southeast Asia: Exploring 
the Linkages. In: Population, Space and Place 22,7: 651-668.
https://doi.org/10.1002/psp.1930

Huguet, Jerrold W. 2003: Can migration avert population decline and ageing in East and 
Southeast Asia? In: Journal of Population Research 20,1: 107-124.
https://doi.org/10.1007/BF03031798

ILO 2024: World employment and social outlook. Geneva: ILO.
Karlinsky, Ariel 2024: International completeness of death registration. In: Demographic 

Research 50: 1151-1170. https://doi.org/10.4054/DemRes.2024.50.38
Kashnitsky, Ilya; de Beer, Joop; van Wissen, Leo 2017: Decomposition of regional convergence 

in population aging across Europe. In: Genus 73,2: 1-25. 
https://doi.org/10.1186/s41118-017-0018-2



The Impact of Migration on Population Ageing in Asia 1990-2020    • 23

Kjærgaard, Søren; Canudas-Romo, Vladimir 2017: Potential support ratios: Cohort versus 
period perspectives. In: Population Studies 71,2: 171-186. 
https://doi.org/10.1080/00324728.2017.1310919

Lee, Ronald; Zhou, Yi 2017: Does Fertility or Mortality Drive Contemporary Population Aging? 
The Revisionist View Revisited. In: Population and Development Review 43,2: 285-301. 
https://doi.org/10.1111/padr.12062

Leasthaghe, Ron; Page, Hillary; Surkyn, Johan 1988: Are Immigrants Substitutes for Births? 
IPD-Working Paper 1988-3. Paris: Interuniversity Programme in Demography.

Liang, Haili; Guo, Zhen; Tuljapurkar, Shripad 2023: Why life expectancy over-predicts crude 
death rate. In: Genus 79,9: 1-14. https://doi.org/10.1186/s41118-023-00188-8

Lischer, Sarah Kenyon 2008: Security and Displacement in Iraq: Responding to the Forced 
Migration Crisis. In: International Security 33,2: 95-119. 
https://doi.org/10.1162/isec.2008.33.2.95

Liu, Peiran; Raftery, Adrian E. 2020: Accounting for uncertainty about past values in probabilistic 
projections of the total fertility rate for most countries. In: The Annals of Applied Statistics 
14,2: 685-705. https://doi.org/10.1214/19-aoas1294

Lutz, Wolfgang; Scherbov, Sergei 2007: The Contribution of Migration to Europe’s Demographic 
Future. Projections for the EU-25 to 2050. IIASA Interim Report. Vienna.

Makaryan, Shushanik; Chobanyan, Haykanush 2014: Institutionalization of Migration Policy 
Frameworks in Armenia, Azerbaijan and Georgia. In: International Migration 52,5: 52-67. 
https://doi.org/10.1111/imig.12163

Manton, Kenneth G.; Gu, XiLiang; Lamb, Vicki L. 2006: Long-Term Trends in Life Expectancy 
and Active Life Expectancy in the United States. In: Population and Development Review 
32,1: 81-105. https://doi.org/10.1111/j.1728-4457.2006.00106.x

McAuliffe, Marie; Triandafyllidou, Anna 2021: World Migration Report 2022. Genova: 
International Organization for Migration.

McDonald, Peter; Kippen, Rebecca 2001: Labor Supply Prospects in 16 Developed Countries, 
2000-2050. In: Population and Development Review 27,1: 1-32. 
https://doi.org/10.1111/j.1728-4457.2001.00001.x

Minnesota Population Center 2020: Integrated Public Use Microdata Series, International: 
Version 7.3 [https://international.ipums.org/international/gis.shtml., 15.12.2024].

Murphy, Michael 2017: Demographic Determinants of Population Aging in Europe since 1850. 
In: Population and Development Review 43,2: 257-283. https://doi.org/10.1111/PADR.12073

Murphy, Michael 2021: Use of Counterfactual Population Projections for Assessing the 
Demographic Determinants of Population Ageing. In: European Journal of Population 37,1: 
211-242. https://doi.org/10.1007/s10680-020-09567-9

Muszyńska, Magdalena M.; Rau, Roland 2012: The Old-Age Healthy Dependency Ratio in 
Europe. In: Journal of Population Ageing 5,3: 151-162. 
https://doi.org/10.1007/s12062-012-9068-6

National Statistics Office of Georgia 2024: Population and Demography. [https://www.geostat.
ge/en/modules/categories/316/population-and-demography, 15.12.2024].

OECD 2022: Pensions at a Glance. Asia/Pacific 2022. Paris: OECD.
Oishi, Nana 2021: Skilled or unskilled?: The reconfiguration of migration policies in Japan. In: 

Journal of Ethnic and Migration Studies 47,10: 2252-2269. 
https://doi.org/10.1080/1369183X.2020.1731984



•    Markus Dörflinger24

Preston, Samuel H.; Himes, Christine; Eggers, Mitchell 1989: Demographic Conditions 
Responsible for Population Aging. In: Demography 26,4: 691-704. 
https://doi.org/10.2307/2061266

Preston, Samuel H.; Vierboom, Yana C. 2021: The Changing Age Distribution of the United 
States. In: Population and Development Review 47,2: 527-539. 
https://doi.org/10.1111/padr.12386

Rau, Roland et al. 2008: Continued Reductions in Mortality at Advanced Ages. In: Population 
and Development Review 34,4: 747-768. https://doi.org/10.1111/j.1728-4457.2008.00249.x

Raymer, James et al. 2022: Estimating international migration flows for the Asia-Pacific region: 
Application of a generation-distribution model. In: Migration Studies 10,4: 631-669.
https://doi.org/10.1093/migration/mnac023

Reslow, Natasja 2017: “Not everything that counts can be counted”: Assessing ‘success’ of EU 
external migration policy. In: International Migration 55,6: 156-169. 
https://doi.org/10.1111/imig.12355

Sanderson, Warren C.; Scherbov, Sergei 2005: Average remaining lifetimes can increase as 
human populations age. In: Nature 435,7043: 811-813. https://doi.org/10.1038/nature03593

Sanderson, Warren C.; Scherbov, Sergei 2007: A new perspective on population aging. In: 
Demographic Research 16: 27-58. https://doi.org/10.4054/DemRes.2007.16.2

Sanderson, Warren C.; Scherbov, Sergei 2008: Rethinking Age and Aging. In: Population 
Bulletin 63,4: 1-20. 

Sanderson, Warren C.; Scherbov, Sergei 2010: Remeasuring aging. In: Science 329,5997: 1287-
1288. https://doi.org/10.1126/science.1193647

Sanderson, Warren C.; Scherbov, Sergei 2013: The Characteristics Approach to the Measurement 
of Population Aging. In: Population and Development Review 39,4: 673-685. 
https://doi.org/10.1111/j.1728-4457.2013.00633.x

Sanderson, Warren C.; Scherbov, Sergei 2020: Choosing between the UN’s alternative views 
of population aging. In: PloS one 15,7: 1-8. https://doi.org/10.1371/journal.pone.0233602

Santis, Gustavo de; Salinari, Giambattista 2023: What drives population ageing? A cointegration 
analysis. In: Statistical Methods & Applications 32,5: 1723-1741.
https://doi.org/10.1007/s10260-023-00713-1

Scherbov, Sergei; Sanderson, Warren C. 2020: New Approaches to the Conceptualization and 
Measurement of Age and Ageing. In: Mazzuco, Stefano; Keilman, Nico (Eds.): Developments 
in Demographic Forecasting. Springer eBook Collection 49. Cham: Springer: 243-258.
https://doi.org/10.1007/978-3-030-42472-5_12

Scherbov, Sergei; Shulgin, S.; Sanderson, Warren C. 2022: World Aging Data Explorer 2022.  
[https://demog.iiasa.ac.at/apps/world.html, 15.12.2024]. 

Shen, Tianyu et al. 2024: The estimation of age and sex profiles for international migration 
amongst countries in the Asia-Pacific region. In: Population, Space and Place 30,1: 1-20.
https://doi.org/10.1002/psp.2716

Siegel, Jacob S.; Hamilton, C. Horace 1952: Some Considerations in the Use of the Residual 
Method of Estimating Net Migration. In: Journal of the American Statistical Association 
47,259: 475-500. https://doi.org/10.1080/01621459.1952.10501186

Skirbekk, Vegard; Loichinger, Elke; Weber, Daniela 2012: Variation in cognitive functioning as 
a refined approach to comparing aging across countries. In: Proceedings of the National 
Academy of Sciences of the United States of America 109,3: 770-774.
https://doi.org/10.1073/pnas.1112173109



The Impact of Migration on Population Ageing in Asia 1990-2020    • 25

Sobotka, Tomáš 2008: Overview Chapter 7: The rising importance of migrants for childbearing in 
Europe. In: Demographic Research 19: 225-248. https://doi.org/10.4054/DemRes.2008.19.9

Spijker, Jeroen J. A. 2023: Combining remaining life expectancy and time to death as a measure 
of old-age dependency related to health care needs. In: International Journal of Health 
Economics and Management 23: 173-187. https://doi.org/10.1007/s10754-022-09328-7

Spoorenberg, Thomas; Schwekendiek, Daniel 2012: Demographic Changes in North Korea: 
1993-2008. In: Population Research and Policy Review 38,1: 133-158. 
https://doi.org/10.1111/j.1728-4457.2012.00475.x

Statistical Committee Republic of Armenia 2024: Statistical Data [https://www.armstat.am/en/, 
15.12.2024].

UN DESA 2000: Replacement Migration. Is it a Solution to Declining and Ageing Populations? 
New York: United Nations.

UN DESA 2019: World Population Ageing 2019. Highlights. New York: United Nations.
UN DESA 2022a: Methodology Report. World Population Prospects 2022. New York: United 

Nations.
UN DESA 2022b: World Population Prospects 2022, Online Edition [https://population.un.org/

wpp/Download, 15.12.2024].
UN DESA 2023: Leaving no one behind in an ageing world. World Social Report 2023. New 

York: United Nations.
Valenta, Marko et al. 2020: Temporary Labour-Migration System and Long-term Residence 

Strategies in the United Arab Emirates. In: International Migration 58,1: 182-197.
https://doi.org/10.1111/imig.12551

Vaupel, James W.; Villavicencio, Francisco; Bergeron-Boucher, Marie-Pier 2021: Demographic 
perspectives on the rise of longevity. In: Proceedings of the National Academy of Sciences 
118,9: 1-10. https://doi.org/10.1073/pnas.2019536118

Weber, Daniela; Loichinger, Elke 2022: Live longer, retire later? Developments of healthy life 
expectancies and working life expectancies between age 50-59 and age 60-69 in Europe. 
In: European Journal of Ageing 19,1: 75-93. https://doi.org/10.1007/s10433-020-00592-5

Weinar, Agnieszka 2014: A Look at Migrations in the Post-Soviet Space – the Case of Eastern 
Europe, South Caucasus and Russian Federation. In: International Migration 52,5: 47-51.
https://doi.org/10.1111/imig.12168

WHO 2023: Global excess deaths associated with COVID-19 (modelled estimates) [https://
www.who.int/data/sets/global-excess-deaths-associated-with-covid-19-modelled-
estimates, 15.12.2024].

Willekens, Frans et al. 2016: International migration under the microscope. In: Science 
352,6288: 897-899. https://doi.org/10.1126/science.aaf6545

Zuo, Wenyun et al. 2018: Advancing front of old-age human survival. In: Proceedings of the 
National Academy of Sciences of the United States of America 115,44: 11209-11214.
https://doi.org/10.1073/pnas.1812337115



•    Markus Dörflinger26

Date of submission: 08.07.2024  Date of acceptance: 24.01.2025

Markus Dörflinger (*). Federal Institute for Population Research (BiB). Wiesbaden, Germany. 
E-mail: markus.doerflinger@bib.bund.de
URL: https://www.bib.bund.de/EN/Institute/Staff/Doerflinger/Doerflinger.html



The Impact of Migration on Population Ageing in Asia 1990-2020    • 27

Appendix

Step-by-step substitution of POADR(t+1) − POADR(t) = α-effect + β-effect (cf. 
equation 5) in section Data and methods) into the α-effects and β-effects by 
component:

1) α  

 [1
2  * (POA(t+1) – POA(t)) * ( 1

PWA(t+1) + 1
PWA(t))] =  

 [1
2  * (CTPOA(t) − DPOA(t) + MPOA(t) + CTLEchange(t)) * ( 1

PWA(t+1) + 1
PWA(t))] =  

 [1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ CTPOA(t)] + 

 [− 1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ DPOA(t)] +  

 [1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ MPOA(t)] +  

 [1
2 ∗ ( 1

PWA(t+1) + 1
PWA(t)) ∗ CTLEchange(t)] 

  

1)

2)2) β  

 [12 ∗ (POA(t+1) + POA(t)) ∗ ( 1
PWA(t+1) −

1
PWA(t))] =  

 [12 ∗ (POA(t + 1) + POA(t)) ∗ (PWA(t)−PWA(t+1)
PWA(t+1)∗PWA(t))] =  

 [12 ∗ (
POA(t+1)+POA(t)
PWA(t+1)∗PWA(t)) ∗ (PWA(t) − PWA(t+1))] =  

 [− 1
2 ∗ (

POA(t+1)+POA(t)
PWA(t+1)∗PWA(t)) ∗ (PWA(t+1) − PWA(t))] =  

 [− 1
2 ∗ (

POA(t+1)+POA(t)
PWA(t+1)∗PWA(t)) ∗ (CTPWA(t) − DPWA(t) + MPWA(t) − CTLEchange(t))] =  

 [− 1
2 ∗

POA(t+1)+POA(t)
PWA(t+1)∗PWA(t) ∗ CTPWA(t)] +  

 [12 ∗
POA(t+1)+POA(t)
PWA(t+1)∗PWA(t) ∗ DPWA(t)] +  

 [− 1
2 ∗

POA(t+1)+POA(t)
PWA(t+1)∗PWA(t) ∗ MPWA(t)] +  

 [12 ∗
POA(t+1)+POA(t)
PWA(t+1)∗PWA(t) ∗ CTLEchange(t)] 



•    Markus Dörflinger28

Tab. A1: Prospective old-age threshold and prospective old-age dependency 
ratio (1990 and 2020)

Prospective old-age threshold Prospective old-age dependency ratio
1990 2020 1990 2020

Central Asia
Kazakhstan 62.8 63.8 11.1 14.1
Kyrgyzstan 62.3 63.9 11.3 7.9
Tajikistan 62.1 63.8 9.2 6.2
Turkmenistan 62.3 63.3 8.9 8.9
Uzbekistan 62.9 62.9 8.3 9.4

Eastern Asia
China 63.5 68.6 9.3 11.7
Hong Kong 68.2 74.7 8.8 10.2
Japan 69.3 74.1 11.7 23.0
Macao 67.9 73.9 7.5 5.5
Mongolia 59.3 65.2 11.3 6.4
North Korea 63.6 66.9 7.3 13.4
South Korea 65.3 72.9 6.8 10.2
Taiwan 65.8 71.7 8.1 10.1

South-Eastern Asia
Brunei Darussalam 64.6 66.2 4.4 6.3
Cambodia 61.4 65.4 8.1 7.5
Indonesia 61.5 63.8 9.7 11.2
Lao 60.1 63.0 10.5 8.2
Malaysia 63.6 66.8 7.1 8.1
Myanmar 60.3 62.5 11.9 12.2
Philippines 62.4 64.3 7.6 8.7
Singapore 65.5 73.3 7.3 5.6
Thailand 68.8 73.8 4.4 7.7
Timor-Leste 57.3 63.0 9.0 10.8
Viet Nam 65.9 68.0 9.2 8.5

Southern Asia
Afghanistan 57.5 62.1 10.2 6.1
Bangladesh 60.1 65.8 10.4 7.6
Bhutan 60.2 64.0 9.4 9.4
India 60.9 65.9 10.5 8.9
Iran 61.9 67.4 8.8 7.8
Maldives 61.4 68.8 8.2 4.2
Nepal 60.1 63.1 10.7 11.4
Pakistan 62.2 63.1 8.7 8.7
Sri Lanka 66.8 67.0 8.2 13.2



The Impact of Migration on Population Ageing in Asia 1990-2020    • 29

Tab. A1: Continuation

Prospective old-age threshold Prospective old-age dependency ratio
1990 2020 1990 2020

Western Asia
Armenia 63.7 66.4 9.5 15.6
Azerbaijan 61.9 65.1 10.2 9.1
Bahrain 64.6 69.1 3.5 2.3
Cyprus 64.5 70.6 15.3 12.1
Georgia 63.8 65.1 15.2 22.2
Iraq 60.5 64.2 10.7 6.3
Israel 68.0 72.4 12.4 9.7
Jordan 63.5 66.7 5.4 4.7
Kuwait 65.8 68.7 2.2 2.9
Lebanon 63.3 68.9 8.4 10.2
Oman 63.4 67.7 5.5 2.8
Palestine 63.0 66.2 5.2 5.3
Qatar 65.2 69.7 1.5 0.7
Saudi Arabia 63.0 68.0 5.6 2.3
Syria 63.4 65.0 6.9 7.5
Turkey 64.7 68.2 8.1 8.6
United Arab Emirates 64.6 69.2 1.6 1.3
Yemen 61.4 62.8 8.8 5.9

Notes: Prospective old-age threshold in years; Prospective old-age dependency ratio in %.
Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger30

Tab. A2: Total change in the prospective old-age dependency ratio and change 
by component (1990-2020) (in percentage points)

Total Cohort Deaths Life expectancy Migration
turnover changes

Central Asia
Kazakhstan 3.0 27.3 -22.1 -1.3 -0.9
Kyrgyzstan -3.4 17.2 -17.8 -1.5 -1.4
Tajikistan -3.0 13.5 -14.7 -1.4 -0.4
Turkmenistan 0.0 17.5 -15.7 -0.8 -1.0
Uzbekistan 1.1 18.9 -17.0 -0.1 -0.7

Eastern Asia
China 2.5 25.0 -16.0 -5.1 -1.5
Hong Kong 1.4 25.8 -16.1 -6.4 -1.9
Japan 11.3 44.5 -26.2 -8.0 1.0
Macao -1.9 15.5 -10.2 -3.6 -3.6
Mongolia -4.9 15.7 -15.6 -5.0 0.0
North Korea 6.1 31.8 -19.9 -5.1 -0.8
South Korea 3.4 23.3 -11.8 -7.0 -1.2
Taiwan 1.9 23.4 -14.3 -5.5 -1.7

South-Eastern Asia
Brunei Darussalam 1.9 12.4 -8.0 -0.6 -1.9
Cambodia -0.5 16.0 -11.9 -3.0 -1.6
Indonesia 1.5 23.4 -18.2 -2.5 -1.2
Lao -2.3 15.6 -18.1 -2.6 2.8
Malaysia 1.0 19.2 -13.2 -2.9 -2.1
Myanmar 0.3 22.5 -19.2 -2.5 -0.6
Philippines 1.1 17.1 -14.9 -1.6 0.4
Singapore -1.7 18.1 -10.2 -5.2 -4.4
Thailand 3.3 18.7 -9.1 -4.2 -2.2
Timor-Leste 1.8 17.3 -17.3 -5.3 7.2
Viet Nam -0.7 16.8 -15.0 -1.5 -1.0

Southern Asia
Afghanistan -4.1 16.2 -15.3 -4.0 -1.1
Bangladesh -2.8 16.3 -14.3 -4.8 0.0
Bhutan 0.1 17.7 -16.7 -3.4 2.4
India -1.6 20.5 -15.8 -5.0 -1.3
Iran -1.0 19.1 -13.3 -5.0 -1.8
Maldives -4.0 14.9 -11.4 -5.2 -2.4
Nepal 0.7 18.7 -17.3 -3.0 2.4
Pakistan 0.0 16.7 -15.6 -1.0 -0.1
Sri Lanka 5.0 25.8 -18.7 -1.5 -0.5



The Impact of Migration on Population Ageing in Asia 1990-2020    • 31

Tab. A2: Continuation

Total Cohort Deaths Life expectancy Migration
turnover changes

Western Asia
Armenia 6.1 29.0 -24.8 -2.9 4.8
Azerbaijan -1.1 22.4 -17.9 -3.3 -2.2
Bahrain -1.2 10.2 -5.1 -1.6 -4.5
Cyprus -3.1 29.4 -22.8 -7.2 -2.5
Georgia 7.0 38.7 -33.7 -1.7 3.7
Iraq -4.4 14.0 -14.1 -3.0 -1.3
Israel -2.7 23.4 -20.8 -4.6 -0.6
Jordan -0.7 15.3 -10.3 -2.5 -3.2
Kuwait 0.7 6.8 -2.6 -0.6 -3.0
Lebanon 1.8 17.5 -13.7 -4.3 2.3
Oman -2.7 10.4 -6.9 -1.8 -4.5
Palestine 0.1 11.9 -9.6 -2.2 0.1
Qatar -0.7 6.8 -2.3 -1.0 -4.3
Saudi Arabia -3.3 9.3 -7.2 -2.0 -3.4
Syria 0.6 13.7 -13.3 -1.1 1.3
Turkey 0.5 19.2 -14.4 -3.3 -1.0
United Arab Emirates -0.3 3.3 -2.2 -0.6 -0.9
Yemen -2.9 11.6 -13.2 -0.9 -0.4

Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger32

Tab. A3: Total change in the prospective old-age dependency ratio and change 
by component (1990-2000) (in percentage points)

Total Cohort Deaths Life expectancy Migration
turnover changes

Central Asia
Kazakhstan 4.6 9.5 -7.4 2.4 0.2
Kyrgyzstan 0.5 6.9 -6.5 0.7 -0.6
Tajikistan 0.1 5.8 -6.1 0.7 -0.3
Turkmenistan 0.2 5.7 -5.2 -0.1 -0.2
Uzbekistan 2.2 6.6 -5.3 0.8 0.0

Eastern Asia
China 0.2 7.2 -4.8 -1.9 -0.3
Hong Kong 0.1 8.5 -4.8 -2.5 -1.1
Japan 1.8 11.8 -6.4 -3.4 -0.2
Macao -1.0 6.0 -3.5 -1.9 -1.6
Mongolia -2.6 4.7 -6.3 -0.8 -0.2
North Korea 6.6 9.5 -4.7 2.6 -0.8
South Korea 0.3 6.1 -3.5 -1.9 -0.4
Taiwan 0.9 7.5 -4.1 -2.1 -0.4

South-Eastern Asia
Brunei Darussalam -0.3 3.0 -2.4 -0.6 -0.4
Cambodia -1.2 4.6 -3.9 -0.8 -1.1
Indonesia 0.7 7.0 -5.2 -0.6 -0.4
Lao -0.8 4.5 -5.3 -0.6 0.5
Malaysia -0.1 4.9 -3.8 -0.6 -0.7
Myanmar -0.8 6.2 -6.2 -0.8 0.0
Philippines -0.1 4.5 -4.3 -0.7 0.4
Singapore -1.7 6.0 -3.7 -2.5 -1.5
Thailand 1.2 4.5 -2.3 0.0 -1.0
Timor-Leste -0.4 6.7 -4.9 -3.6 1.4
Viet Nam -1.1 4.9 -4.6 -1.0 -0.3

Southern Asia
Afghanistan -2.2 5.1 -4.4 -2.3 -0.6
Bangladesh -3.8 4.8 -4.9 -3.6 -0.1
Bhutan 0.5 6.3 -4.4 -1.2 -0.2
India -1.3 6.1 -5.3 -1.7 -0.4
Iran -1.1 4.8 -4.3 -1.3 -0.4
Maldives -0.6 5.3 -4.3 -2.2 0.5
Nepal -1.3 5.5 -5.2 -1.5 -0.2
Pakistan -0.1 5.0 -4.7 0.0 -0.3
Sri Lanka 3.7 5.5 -4.8 2.8 0.2



The Impact of Migration on Population Ageing in Asia 1990-2020    • 33

Tab. A3: Continuation

Total Cohort Deaths Life expectancy Migration
turnover changes

Western Asia
Armenia 6.3 10.9 -6.9 0.1 2.2
Azerbaijan 1.9 9.6 -6.4 -0.3 -1.0
Bahrain -0.3 3.1 -1.7 -0.7 -1.0
Cyprus -2.5 8.5 -8.7 -2.5 0.1
Georgia 5.7 15.0 -10.3 0.4 0.5
Iraq -3.2 3.5 -4.9 -1.4 -0.4
Israel -0.4 7.6 -7.1 -1.2 0.2
Jordan -0.3 3.9 -2.8 -0.5 -0.8
Kuwait -0.5 1.7 -0.8 -0.2 -1.1
Lebanon -0.4 5.8 -4.3 -1.7 -0.2
Oman -1.5 3.1 -2.6 -0.9 -1.1
Palestine -0.2 3.2 -2.6 -0.5 -0.3
Qatar 0.0 2.0 -0.7 -0.1 -1.1
Saudi Arabia -1.1 3.1 -2.8 -0.7 -0.6
Syria 0.0 3.9 -3.7 0.0 -0.1
Turkey 0.1 5.3 -4.5 -0.4 -0.3
United Arab Emirates -0.5 1.3 -0.7 -0.2 -0.9
Yemen -1.3 3.7 -4.4 -0.6 -0.1

Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger34

Tab. A4: Total change in the prospective old-age dependency ratio and change 
by component (2000-2010) (in percentage points)

Total Cohort Deaths Life expectancy Migration
turnover changes

Central Asia
Kazakhstan -3.2 6.9 -8.6 -1.2 -0.3
Kyrgyzstan -3.9 2.7 -6.2 -0.4 0.0
Tajikistan -2.6 1.8 -4.6 -0.4 0.5
Turkmenistan -2.0 3.2 -4.9 -0.4 0.0
Uzbekistan -2.1 3.3 -5.3 -0.3 0.2

Eastern Asia
China 0.2 6.7 -4.9 -1.5 -0.1
Hong Kong 1.4 8.5 -4.9 -1.9 -0.4
Japan 5.0 14.7 -7.9 -2.1 0.3
Macao -1.4 4.1 -3.0 -1.0 -1.6
Mongolia -1.4 3.9 -4.7 -1.0 0.5
North Korea 1.4 12.6 -7.5 -3.5 -0.2
South Korea 1.5 8.2 -3.7 -2.8 -0.2
Taiwan -0.4 7.0 -4.5 -2.1 -0.7

South-Eastern Asia
Brunei Darussalam 0.3 3.2 -2.2 -0.2 -0.6
Cambodia -1.1 3.6 -3.2 -1.3 -0.1
Indonesia 0.4 6.9 -5.6 -0.8 -0.2
Lao -1.2 4.0 -5.8 -0.9 1.5
Malaysia 0.1 5.2 -3.8 -1.0 -0.3
Myanmar -0.6 6.2 -6.1 -0.6 -0.1
Philippines 0.2 4.6 -4.5 -0.3 0.4
Singapore -0.4 5.1 -2.8 -1.4 -1.3
Thailand 0.8 6.2 -2.8 -2.0 -0.7
Timor-Leste 3.2 5.8 -5.2 -1.5 4.1
Viet Nam -0.5 4.2 -4.4 -0.3 0.1

Southern Asia
Afghanistan -0.9 4.1 -3.8 -0.9 -0.2
Bangladesh 1.4 4.7 -4.1 0.5 0.3
Bhutan -0.5 5.1 -5.6 -1.3 1.3
India -0.2 5.6 -4.9 -0.7 -0.2
Iran -1.1 4.5 -3.8 -1.3 -0.4
Maldives -2.3 4.1 -3.3 -2.5 -0.6
Nepal 0.9 5.8 -5.0 -0.7 0.8
Pakistan -0.3 4.8 -4.6 -0.4 -0.2
Sri Lanka -0.6 7.8 -6.8 -1.2 -0.3



The Impact of Migration on Population Ageing in Asia 1990-2020    • 35

Tab. A4: Continuation

Total Cohort Deaths Life expectancy Migration
turnover changes

Western Asia
Armenia -0.9 6.4 -8.9 -1.3 2.9
Azerbaijan -3.1 3.8 -5.8 -0.9 -0.2
Bahrain -1.2 2.4 -1.3 -0.6 -1.7
Cyprus -2.1 8.3 -6.7 -2.7 -1.0
Georgia -0.3 10.2 -12.1 -1.2 2.8
Iraq -0.5 3.9 -4.1 0.1 -0.3
Israel -1.7 6.8 -6.4 -2.2 0.1
Jordan 0.4 3.9 -2.7 -0.6 -0.2
Kuwait 0.0 1.9 -0.9 -0.2 -0.9
Lebanon -1.1 4.7 -3.7 -2.0 -0.1
Oman -0.8 2.5 -2.0 -0.5 -0.8
Palestine 0.1 3.1 -2.6 -0.6 0.2
Qatar -0.8 1.5 -0.5 -0.3 -1.4
Saudi Arabia -1.7 2.3 -2.1 -0.6 -1.4
Syria -1.4 3.3 -3.2 -1.0 -0.5
Turkey 0.1 5.8 -4.5 -0.9 -0.2
United Arab Emirates -0.9 0.6 -0.4 -0.1 -1.1
Yemen -1.2 3.1 -3.6 -0.6 -0.1

Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger36

Tab. A5: Total change in the prospective old-age dependency ratio and change 
by component (2010-2020) (in percentage points)

Total Cohort Deaths Life expectancy Migration
turnover changes

Central Asia
Kazakhstan 1.5 11.5 -7.1 -2.2 -0.6
Kyrgyzstan 0.0 6.6 -4.7 -1.3 -0.6
Tajikistan -0.4 4.9 -3.8 -1.1 -0.3
Turkmenistan 1.9 7.0 -4.3 -0.2 -0.6
Uzbekistan 1.1 7.1 -5.0 -0.4 -0.6

Eastern Asia
China 2.0 9.9 -5.5 -1.5 -0.9
Hong Kong 0.0 7.5 -5.4 -1.7 -0.4
Japan 4.6 17.0 -11.2 -2.1 0.9
Macao 0.5 4.2 -2.8 -0.7 -0.3
Mongolia -0.9 5.5 -3.8 -2.4 -0.2
North Korea -1.9 8.9 -7.1 -3.9 0.1
South Korea 1.5 8.6 -4.3 -2.1 -0.6
Taiwan 1.4 8.0 -5.0 -1.1 -0.5

South-Eastern Asia
Brunei Darussalam 2.0 5.0 -2.6 0.1 -0.6
Cambodia 1.7 5.7 -3.2 -0.5 -0.3
Indonesia 0.4 7.7 -5.9 -0.9 -0.5
Lao -0.3 5.3 -5.3 -0.8 0.5
Malaysia 1.1 6.5 -3.9 -0.8 -0.7
Myanmar 1.6 9.0 -6.0 -1.0 -0.4
Philippines 0.9 6.3 -4.6 -0.4 -0.3
Singapore 0.4 4.9 -2.6 -1.0 -0.9
Thailand 1.3 6.6 -3.2 -1.7 -0.4
Timor-Leste -1.0 4.8 -6.7 -0.8 1.7
Viet Nam 0.8 6.1 -4.4 -0.2 -0.6

Southern Asia
Afghanistan -1.1 3.2 -3.4 -0.5 -0.4
Bangladesh -0.4 5.7 -4.2 -1.6 -0.3
Bhutan 0.1 5.8 -5.9 -0.8 1.0
India -0.1 7.0 -4.5 -2.1 -0.6
Iran 1.2 6.8 -3.5 -1.5 -0.6
Maldives -1.1 2.9 -2.5 -0.5 -1.1
Nepal 1.1 6.5 -5.8 -0.7 1.0
Pakistan 0.4 5.0 -4.5 -0.4 0.2
Sri Lanka 1.9 10.9 -6.2 -2.5 -0.2



The Impact of Migration on Population Ageing in Asia 1990-2020    • 37

Tab. A5: Continuation

Total Cohort Deaths Life expectancy Migration
turnover changes

Western Asia
Armenia 0.7 11.2 -9.2 -1.9 0.5
Azerbaijan 0.2 7.8 -5.1 -1.6 -0.9
Bahrain 0.4 2.3 -1.1 -0.2 -0.7
Cyprus 1.5 9.8 -5.9 -1.5 -0.9
Georgia 1.7 14.9 -13.1 -1.3 1.1
Iraq -0.7 4.5 -3.6 -1.3 -0.3
Israel -0.5 6.7 -5.6 -0.8 -0.8
Jordan -0.8 3.9 -2.6 -0.7 -1.4
Kuwait 1.1 3.1 -0.9 -0.2 -0.8
Lebanon 3.3 6.2 -4.3 -0.4 1.9
Oman -0.4 3.0 -1.6 -0.3 -1.5
Palestine 0.2 3.4 -2.6 -0.7 0.0
Qatar 0.0 1.1 -0.3 -0.2 -0.5
Saudi Arabia -0.5 2.1 -1.4 -0.4 -0.8
Syria 2.0 4.5 -3.8 0.1 1.1
Turkey 0.3 6.6 -4.4 -1.5 -0.4
United Arab Emirates 1.1 0.5 -0.3 -0.1 1.0
Yemen -0.4 2.9 -3.3 0.2 -0.1

Source: UN DESA 2022b, own calculations.



•    Markus Dörflinger38

Tab. A6: Mean annual net migration rates 1990-2019, 1990-1999, 2000-2009, 
2010-2019 (per 1,000 population)

1990-2019 1990-1999 2000-2009 2010-2019

Central Asia
Kazakhstan -6.3 -19.1 0.0 -0.1
Kyrgyzstan -5.9 -6.7 -5.4 -6.2
Tajikistan -5.9 -12.0 -3.8 -2.7
Turkmenistan -1.5 -0.7 -2.0 -1.7
Uzbekistan -1.5 -2.1 -1.5 -1.0

Eastern Asia
China -0.3 -0.5 -0.4 -0.1
Hong Kong 4.3 8.1 3.4 2.0
Japan 0.9 0.3 1.0 1.2
Macao 15.0 11.5 21.1 12.7
Mongolia -2.2 -3.9 -3.8 0.7
North Korea -0.1 -0.1 -0.1 -0.1
South Korea -0.2 -3.7 -0.6 3.9
Taiwan -0.2 -2.0 0.1 1.3

South-Eastern Asia
Brunei Darussalam 2.3 4.7 3.3 -0.9
Cambodia 0.6 6.1 -1.0 -3.0
Indonesia -0.2 0.0 -0.4 -0.2
Lao -3.8 -3.4 -5.3 -3.0
Malaysia 5.5 5.1 8.7 3.1
Myanmar -2.2 -1.6 -3.4 -1.7
Philippines -1.5 -1.6 -2.3 -0.9
Singapore 15.5 19.8 17.5 10.8
Thailand 1.3 1.7 1.6 0.8
Timor-Leste -5.9 -8.7 -6.5 -2.5
Viet Nam -0.5 0.2 -1.6 0.0

Southern Asia
Afghanistan 8.4 25.2 -0.5 1.0
Bangladesh -3.1 -2.8 -5.1 -1.7
Bhutan -4.3 -17.1 4.4 -0.9
India -0.2 0.0 -0.4 -0.2
Iran 1.5 0.5 2.7 1.8
Maldives 8.9 -1.9 7.8 22.6
Nepal -5.7 -2.0 -8.5 -8.8
Pakistan -3.2 -0.6 -1.8 -7.5
Sri Lanka -3.0 -3.7 -1.2 -3.8



The Impact of Migration on Population Ageing in Asia 1990-2020    • 39

Tab. A6: Continuation

1990-2019 1990-1999 2000-2009 2010-2019

Western Asia
Armenia -13.3 -19.9 -11.3 -9.6
Azerbaijan -0.6 -3.8 1.6 0.2
Bahrain 17.9 11.6 39.2 9.9
Cyprus 8.9 10.2 12.5 4.9
Georgia -14.0 -26.3 -12.8 -4.6
Iraq 0.4 0.4 -3.9 5.2
Israel 4.3 9.2 2.4 1.5
Jordan 13.1 8.7 6.3 27.8
Kuwait 12.2 -8.3 21.5 33.3
Lebanon 0.1 -0.2 0.7 6.5
Oman 9.6 2.1 -3.8 41.3
Palestine -4.0 1.0 -7.5 -5.8
Qatar 47.6 18.7 90.4 48.6
Saudi Arabia 5.1 4.0 9.8 4.6
Syria -5.7 -1.0 5.1 -28.6
Turkey 0.1 -1.1 -0.6 2.5
United Arab Emirates 40.3 37.1 89.3 -1.6
Yemen -2.3 -2.7 -3.2 -1.1

Source: UN DESA 2022b, own calculations.



Published by
Federal Institute for Population Research 
(BiB)
65180 Wiesbaden / Germany

 2025

Editor 
Prof. Dr. Roland Rau
Prof. Dr. Heike Trappe

Managing Editor
Dr. Katrin Schiefer

Editorial Assistant
Beatriz Feiler-Fuchs
Wiebke Hamann

Layout
Beatriz Feiler-Fuchs

E-mail: cpos@bib.bund.de

Comparative Population Studies

www.comparativepopulationstudies.de

ISSN: 1869-8980 (Print) – 1869-8999 (Internet)

Scientific Advisory Board
Kieron Barclay (Stockholm)
Ridhi Kashyap (Oxford)
Anne-Kristin Kuhnt (Rostock)
Mathias Lerch (Lausanne)
Eleonora Mussino (Stockholm)
Natalie Nitsche (Canberra)
Alyson van Raalte (Rostock)
Pia S. Schober (Tübingen)
Sergi Vidal (Barcelona)
Rainer Wehrhahn (Kiel)