Pattern and trends of population age distribution

Pattern and trends of population age distribution

SÓLO PARA PARTICIPANTES

Gobierno de Argentina, Ministerio de Desarrollo Social de la Nación

Instituto Nacional de Servicios Sociales para Jubilados y Pensionados

DOCUMENTO DE REFERENCIA

10 de Octubre de 2005

SOLO INGLÉS

Comisión Económica para América Latina y el Caribe (CEPAL)

CELADE- División de Población

Reunión de Gobiernos y Expertos sobre Envejecimiento de Países de América del Sur

Con la colaboración del Fondo de Población de las Naciones Unidas (UNFPA), Organización Panamericana de la Salud (OPS), Banco Interamericano de Desarrollo (BID) y Organización Internacional del Trabajo

(OIT)

Buenos Aires, Argentina, 14 al 16 de noviembre de 2005

THE DIVERSITY OF CHANGING POPULATION AGE

STRUCTURES IN THE WORLD

Este documento fue preparado por PAULO SAAD de la División de Población de las Naciones Unidas. Las opiniones expresadas en este documento, que no ha sido sometido a revisión editorial, son de la exclusiva responsabilidad del autor y pueden no coincidir con las de la Organización. Se prohíbe citar sin autorización del autor.

THE DIVERSITY OF CHANGING POPULATION AGE

STRUCTURES IN THE WORLD

Population Division

Department of Economic and Social Affairs

United Nations Secretariat

2

T

HE DIVERSITY OF CHANGING POPULATION AGE STRUCTURES IN THE WORLD

The world population is in the midst of an unprecedented transformation brought about by the transition from a regime of high mortality and high fertility to one of low mortality and low fertility. This demographic transition is responsible for the rapid and accelerating growth that the world population experienced in the twentieth century as well as for the slowing down of that growth and for the changes in the age distribution associated with those developments. Indeed, the demographic transition starts usually with a reduction of mortality, which results in longer survival, particularly of children who typically benefit the most from the reduction of the very high risks of death that they experience when mortality is high. As a consequence population growth accelerates and the proportion of children in the population increases, leading to a rejuvenation of the population’s age structure. Partly in response to these changes, fertility begins to decrease because parents realize that they can have fewer children to ensure the survival of the number they desire. Sustained reductions of fertility slow down population growth and produce eventually reductions of the proportion of children in the population thus triggering the process of population ageing. As time elapses, if the reductions of fertility and mortality continue, they reinforce the ageing process because, over time, sustained fertility decline leads not only to decreasing proportions of children but also of young people and eventually of adults of working age. Furthermore, increases in longevity have generally the effect of accelerating the growth of the proportion of older persons more than those of young people or adults

Thus, in terms of the effects of the demographic transition on population age structures, one can distinguish three distinct stages. During the first, there is a rejuvenation of the age distribution as the proportion of children increases. During the second, triggered by fertility reductions, the proportion of children begins to decline while the proportion of adults and older persons rise. During the third stage, reached usually after lengthy periods of fertility and mortality decline, the proportions of both children and adults of working age decline and only the proportion of older persons rises.

During the second stage of the transition, adults of working age constitute a significantly larger proportion of the total population than during the first stage of the transition, so that the number of adults of working age per dependent (that is, children and older persons) increases for a certain period until it reaches a maximum. During that period, a population is optimally placed to benefit from economically productive investment because its levels of economic dependency are low and there are relatively more potential workers to support persons in the non-productive ages (children and the older population). The terms “demographic dividend”, “demographic bonus” or “demographic window of opportunity” have been coined to describe this stage of the transition and allude to the possibilities that it presents for raising a country’s rate of economic growth per capita and its standard of living.

However, the benefits associated with the “demographic window of opportunity” are not automatic.

For instance, the second stage of the demographic transition usually entails a rapidly growing population of young people (aged 15 to 24) who need to be educated and provided with gainful employment in order to become assets for society. The demographic bonus may provide an opportunity for speeding development, but realizing those benefits depends on the adoption of macroeconomic policies that promote productive investment, increase employment opportunities and in general ensure a stable social and economic environment propitious for sustained economic growth and sustainable development.

Although the period of the demographic bonus can last several decades, if fertility reductions are sustained it eventually gives way to the third stage of the transition, where the proportion of adults of working age ceases to rise and only the proportion of older persons continues increasing. This period of rapid population ageing poses new challenges for the adaptation of society to an unprecedented situation

3

and requires the development of sound public policy to facilitate the adjustments that will be necessary in a variety of spheres, including the provision of health care and old-age support. The need for such adjustments raises questions of intergenerational equity in meeting the needs of persons at different stages of life and of the roles of the State, the private sector and the family in providing such support. Different approaches to solving those problems are likely to affect the distribution of wealth, opportunities and burdens that cut across age and gender groups within societies.

Today the major areas of the world are at different stages of the demographic transition. Europe is well into the third stage of the transition and its population, which is already the oldest in the world, is expected to age rapidly in the foreseeable future. Northern America and Australia/New Zealand also find themselves in the third stage of the transition but because their fertility levels have not fallen as low as those of Europe, they are expected to experience a somewhat slower ageing process. Asia and Latin

America and the Caribbean find themselves in the second stage of the transition and are still in time to benefit from the demographic bonus. However, because these two regions experienced on average fairly rapid fertility reductions, they are expected to age more rapidly than Europe or Northern America did in the past. Lastly, Africa has only recently embarked on the second stage of the transition and still has a very young population with high dependency levels. Furthermore, because Africa is the major area most affected by the HIV/AIDS epidemic, its transition to low mortality has been interrupted and it is not clear whether the incipient fertility reductions experienced by countries in the region will accelerate or not over the short term. Even assuming that fertility reductions proceed at a moderate pace, Africa is expected to continue being the major area with the youngest population well into the 21 st

century.

This paper presents an overview of trends in population ageing, thus providing the demographic basis for the further examination of the social, economic and policy implications of the changes expected.

Most of the data presented here are derived from the 2004 Revision of World Population Prospects, the most recent version of the official United Nations population estimates and projections (United Nations,

2005a). In what follows, trends are generally discussed in terms of the following age groups: children (the population aged 0 to 14 years); young persons or youth (persons aged 15 to 24 years); adults of working age (aged 15 to 59); older persons (persons aged 60 years or over), and the oldest-old (those aged 80 years or over).

A. T

HE CHANGING AGE DISTRIBUTION OF THE WORLD AND THE DEVELOPMENT GROUPS

As described above, the different stages of the demographic transition give rise to various changes of the age distribution over time. At the world level, the population in 1950 was relatively young, having 34 per cent of its members under age 15 and barely 8 per cent aged 60 or over (table 1). Between 1950 and

1975, as mortality decline accelerated, particularly in the less developed regions, both the proportion under age 15 and that aged 60 or over increased, to reach 37 per cent and about 9 per cent respectively.

Overall, therefore, the population of the world became slightly younger from 1950 to 1975. But after

1975, as fertility reductions in the developing world accelerated, the proportion of children at the world level began to decrease, so that by 2005 the population under age 15 accounted for just 28 per cent of the total. Given that fertility at the world level started declining in the 1970s, by 2000 there had also been a slight reduction in the proportion of the population aged 15-24, from 19 per cent in 1975 to 18 per cent in

2005. However, as expected, the proportion aged 25-59 had risen markedly, passing from 36 per cent in

1975 to nearly 44 per cent in 2005. These trends suggests that the world population is already well into the period in which the demographic bonus can be reaped. Furthermore, if fertility and mortality trends continue as projected in the medium variant of the official United Nations population projections, the bonus will persist until at least 2025, since still by then it is projected that the proportion of the population aged 25-59 will continue to increase, albeit slightly.

4

T

ABLE

1. P

OPULATION BY MAJOR AGE GROUP AND PERCENTAGE DISTRIBUTION BY AGE GROUP FOR THE WORLD AND

THE DEVELOPMENT GROUPS

, 1950, 1975, 2005, 2025

AND

2050

Age group

1950

Population (millions)

1975 2005 2025 2050 1950 1975

Percentage

2005 2025 2050

0-14

15-24

25-59

60-79

80+

Total

864 1 498 1 821 1 909

459 757 1 159 1 211

991 1 469 2 812 3 593

192 318 586 1 032

14 31 87 160

2 519 4 074 6 465 7 905

World

1 833 34.3 36.8 28.2 24.2

1 225 18.2 18.6 17.9 15.3

4 051 39.3 36.1 43.5 45.4

1 574 7.6 7.8 9.1 13.1

394 0.5 0.8 1.3 2.0

9 076 100.0 100.0 100.0 100.0

20.2

13.5

44.6

17.3

4.3

100.0

0-14

15-24

25-59

60-79

80+

Total

222

138

357

87

254

176

456

143

9 18

813 1 047

206

165

596

200

196

140

570

275

More developed regions

193

133

510

27.4

17.0

43.9

24.2

16.8

43.5

17.0

13.7

49.2

15.7

11.2

45.7

284 10.7 13.7 16.5 22.0

44 68 116 1.0 1.8 3.7 5.4

1 211 1 249 1 236 100.0 100.0 100.0 100.0

15.6

10.8

41.2

23.0

9.4

100.0

0-14

15-24

25-59

60-79

80+

Total

642 1 244

321 581

634 1 014

105

5

175

13

1 707 3 027

Less developed regions

1 615 1 713 1 639 37.6 41.1 30.7 25.7

994 1 071 1 091 18.8 19.2 18.9 16.1

2 216 3 023 3 541 37.1 33.5 42.2 45.4

386

42

757

92

1 290

278

6.1

0.3

5.8

0.4

7.3

0.8

11.4

1.4

5 253 6 656 7 840 100.0 100.0 100.0 100.0

20.9

13.9

45.2

16.5

3.6

100.0

Source: World Population Prospects: The 2004 Revision, Datasets in Excel and PDF Formats, Extended

Dataset, United Nations, 2005.

These trends at the world level are mainly the result of the average trends for the less developed regions, where the age distribution also showed an increase in the proportion of children and youth between 1950 and 1975, followed by a decline in both proportions between 1975 and 2005. But just as at the world level, this decline is compensated by a marked increase in the proportion aged 25-59 which is expected to continue rising until at least 2025 and could remain almost unchanged until 2050 (at close to

45 per cent). In addition, the proportion aged 60 or over in the less developed regions has been rising steadily and is expected to pass from 8 per cent in 2005 to 13 per cent in 2025 and might reach nearly 20 per cent by 2050. That is, the ageing of the population of less developed regions is expected to accelerate, particularly after 2025.

In contrast with the less developed regions, whose population is still fairly young, the more developed regions have an older population. Already today, 20 per cent of the population in the more developed regions is aged 60 years or over, a proportion that might be reached by the less developed regions only by 2050. However, because of the very low fertility that the more developed regions have had since the 1980s, their age distribution today is somewhat more advantageous that the one projected for less developed regions in 2050, in the sense that the proportion of persons aged 25-59 in more

5

developed regions today is 49 per cent or 4 percentage points higher than the proportion projected for less developed regions in 2050.

Another way of assessing the implications of population distribution for economic and social development is to consider the number of persons aged 25-59 per 100 persons in other age groups. This ratio is a rough indicator of the number of persons in the economically productive ages to the number of persons of other ages who are more likely not to be productive, either because they are studying or because they are no longer economically active.

1

Table 2 shows the evolution of this ratio between 1950

and 2050 for the world and the development groups. The ratios indicate that since 1950, the more developed regions have had a considerably more advantageous ratio than the less developed regions.

Especially today, there are 97 persons aged 25-59 in more developed regions to every 100 persons in the other age groups, whereas in the less developed regions that ratio, which is at an all time high, is just 73.

Over the next 20 years, the situation in the less developed regions continues improving while the ratio for the more developed regions declines, but by 2025 both sets of countries are expected to have similar overall ratios. After 2025, the ratio for the more developed regions declines markedly while that for the less developed regions remains almost unchanged. This measure corroborates therefore that the less developed regions, as a whole, still have the opportunity to reap the benefits of the demographic bonus and some time to adapt to the rapid population ageing that they will experience once the bonus ends.

T

ABLE

2. N

UMBER OF PERSONS AGED

25-59

PER

100

PERSONS IN THE OTHER AGE

GROUPS FOR THE WORLD AND THE DEVELOPMENT GROUPS

, 1950-2050

World

1950 1975 2005 2025 2050

65 56 77 83 81

78 77 97 84 70

59 50 73 83 82

Source: World Population Prospects: The 2004 Revision, Datasets in

Excel and PDF Formats, Extended Dataset, United Nations, 2005.

It should be noted that although the proportion of the population aged 25-59 is expected to remain nearly constant in the less developed regions between 2005 and 2050, the age distribution of those potential workers is also expected to become older, as illustrated in figure 1. Thus, whereas in 1950 the age distribution of the working-age population in the less developed regions resembled a pyramid, with a broad base at the younger working ages and decreasing numbers at higher ages, by 2050 the expected distribution is expected to become closer to a rectangle, although the younger cohorts in the less developed regions still tend to be somewhat larger than the older cohorts. In the more developed regions, the distribution of the group aged 25-59 today is closer to an inverted pyramid, with the older cohorts being larger than the younger ones, and the expected distribution by 2050 also becomes considerably more rectangular. Furthermore, whereas in the less developed regions the population continues to increase until 2050 for every age group above age 10, the population of the more developed regions is smaller in

2050 than it is expected to be in 2025 for every age group above age 10.

1

The ratio presented here differs from that normally used which considers age group 15-59 to represent those ages where economic activity is most prevalent and 0-14 together with 60+ as the age groups where people are more likely to be economically dependent. The findings using these groups would not be fundamentally different.

Furthermore, other types of dependency ratios will be considered below.

6

Figure 1. Age and sex distribution of the population, more developed and less developed regions,

1950, 2005 and 2050

More Developed Regions

Less Developed Regions

100

90

Males Females

100

90

80

Males Females

80

70

70

60

60

50

50

40

40

30

20

10

1950

2005

2050

30

20

10

0

15 10 5 0 5 10 15

0

60 50 40 30 20 10 0 10 20 30 40 50 60

Population (millions per year of age)

Population (millions per year of age)

Source: World Population Prospects: The 2004 Revision, Datasets in Excel and PDF Formats, Extended

Dataset, United Nations, 2005.

In sum, the world population as a whole is today in a relatively favourable position regarding the size of its population of working age relative to that of children and older persons. Furthermore, although the more developed regions are more advanced in the process of population ageing than the less developed regions, they are still in a relatively advantageous situation because they have a fairly high proportion of persons in the economically active ages and a moderate proportion of older persons (20 per cent). Their main concern, therefore, should be to make provisions for the rapid ageing that lies ahead. For the less developed regions as a whole, the situation is also propitious and the expectation is that it will improve further over the short term as fertility continues to decrease. However, this development will help accelerate the ageing process so that by mid-century the less developed regions are likely to be in a position similar to that of more developed countries today.

B. P

OPULATION AGEING IN MAJOR AREAS

The overall trends described above mask important differences among major areas and countries, resulting from variations in the timing of the demographic transition in each of them. Table 3 displays selected indicators of the changing age distribution in each major area and figures 2 and 3 show trends over time. Clearly Africa has today the youngest age distribution, with 41 per cent of the population under age 15 and just about 5 per cent of the population aged 60 years or over. In sharp contrast, Europe has a much older population, with just 16 per cent under age 15 and 21 per cent aged 60 or over. Therefore, in

Europe the number of older persons has already surpassed the number of children.

7

0-14

15-24

25-59

60-79

80+

Total

0-14

15-24

25-59

60-79

80+

Total

0-14

15-24

25-59

60-79

80+

Total

25-59

60-79

80+

Total

0-14

15-24

25-59

60-79

80+

Total

15-24

25-59

60-79

80+

Total

0-14

15-24

T

ABLE

3. P

OPULATION BY MAJOR AGE GROUP AND PERCENTAGE DISTRIBUTION BY AGE GROUP AND

MAJOR AREA

, 1950, 1975, 2005, 2025

AND

2050

Age group 1950

Population (millions)

1975 2005 2025

Proportion by age group (percentage)

2050 1950 1975 2005 2025 2050

0-14

Africa

94 187 376 495 556 42 45 41 37 29

42 79 189 270 350 19 19 21 20 18

76 129 294 494 838 34 31 32 37 43

11

1

20

1

44

4

78

7

173

20

5

0

5

0

5

0

6

1

9

1

224 416 906 1 344 1 937 100 100 100 100 100

Asia

509 949 1 086 1 067 954 36 40 28 23 18

263 455 712 701 650 19 19 18 15 12

529 832 1 743 2 254 2 382 38 35 45 48 46

90 148 327 623 996 6 6 8 13 19

4 11 38 83 235 0 0 1 2 5

1 396 2 395 3 905 4 728 5 217 100 100 100 100 100

67 133 168 163 141 40 41 30 23 18

31

59 105 238 323 354 35 33 42 46 45

9

1

167 322 561 697 783 100 100 100 100 100

47

26

19

2

63 106 110

19

2

62

46

43

7

Latin America and the Caribbean

15

99 19 20 19 16 13

86 148

40

6

0

6

1

8 12 19

1 2 5

78 101 161 173 192 45 41 49 45 44

30

5

68

47

44

12

71

50

78

16

Northern America

75

52 15 19 14 13 12

85

33

27

11

1

25

12

2

20

13

4

18

20

4

17

19

8

172 243 331 388 438 100 100 100 100 100

143 160 115 104

95 111 101 75

Europe

98 26 24 16 15 15

67 17 16 14 11 10

243 294 361 330 263 44 43 50 47 40

60

6

99

12

125

26

161

37

163

63

11

1

15

2

17

4

23

5

25

10

547 676 728 707 653 100 100 100 100 100

4

2

6

1

0

13

7

4

8

2

0

21

8

5

15

4

1

33

9

5

19

7

1

41

Oceania

9 30 31 25 21 18

6 16 18 16 13 12

21 43 40 46 45 45

9 10 10 11 17 18

3 1 1 3 3 7

48 100 100 100 100 100

Source: World Population Prospects: The 2004 Revision, Datasets in Excel and PDF

Formats, Extended Dataset, United Nations, 2005.

8

Figure 2. Percentage distribution of the population by broad age group and major area, 1950-2050

Africa Asia Europe

100

60-79

80+

100

80

60-79

80+

100

80

60-79

80+

100

80

80

25-59

25-59

60

60 60

60

25-59

15-24 15-24

40 40 40

15-24

40

0-14

20

0-14

20

0-14

20

20

1950 1970 1990 2010

2005

2030

Year

2050

0

1950 1970 1990 2010

2005

2030

Year

2050

0

1950 1970 1990 2010

2005

2030

Year

0

2050

100

80

Latin America and the Caribbean

60-79

80+

100

80

Northern America

60-79

80+

100

80

Oceania

80+

60-79

100

80

25-59

60

60

60

25-59

60

25-59

15-24

40 40 40 40

15-24

15-24

20 20

20 20

0-14 0-14 0-14

0

1950 1970 1990 2010

2005

2030

Year

2050

0

1950 1970 1990 2010

2005

2030

Year

2050

0

1950 1970 1990 2010

2005

2030

Year

0

2050

Source: World Population Prospects: The 2004 Revision, Datasets in Excel and PDF Formats, Extended Dataset, United

Nations, 2005.

Northern America is the second oldest area, with 20 per cent of the population under age 15 and 17 per cent aged 60 years or over. Oceania, where the developed countries of Australia and New Zealand account for most of the population, exhibits the next oldest population, with 25 per cent of children and

14 per cent of older persons. Asia and Latin America and the Caribbean follow. Both have about 9 per cent of the population aged 60 years or over, but Latin America and the Caribbean have a higher proportion of children than Asia (30 per cent vs. 28 per cent).

Although all major areas are expected to experience further ageing of their populations over the next

45 years, large differences in age structure will likely persist until 2050. By then, the proportion aged 60 years or over is likely to range from 10 per cent in Africa to 35 per cent in Europe. In all the other major areas, about a quarter of the population is expected to be 60 or over. Similarly, except for Africa and

Europe, where the proportion of children is expected to be 29 per cent and 15 per cent, respectively, that of all other areas is expected to be between 17 and 18 per cent. This implies that in all major areas except

Africa and Europe, the proportion of the population aged 15-59 is expected to be similar, ranging from 56 to 58 per cent. Europe would have by 2050 a much lower proportion in those ages (50 per cent), while

Africa would have 61 per cent.

One important development expected over the next century is the continued ageing of the older population. That is, the proportion of persons aged 80 or over is expected to grow rapidly. Thus, whereas in 2005 that proportion amounted to just over 1 per cent of the world population, by 2050 it is expected to reach over 4 per cent. In Europe and Northern America, where the oldest old constitute today about 4 per cent of the population, they are expected to account for 10 per cent and 8 per cent of their respective

9

Figure 3. Evolution of the proportion of the population in major age groups for the world and major areas, 1950-2050

World

70

60

50

Ages 15-59

40

30

20

10

0

1950

Children

(< 15 years)

1970 1990

Year

2010

Older persons

(age 60+)

2030 2050

70

60

50

40

30

20

10

0

1950

Africa

70

60

50

40

30

20

10

0

1950 1970 1990

Year

2010 2030

More developed regions

1970 1990

Year

2010 2030 2050

2050

Asia

50

40

30

70

60

20

10

0

1950 1970 1990

Year

2010 2030 2050

Less developed regions

70

60

50

40

30

20

10

0

1950 1970 1990

Year

2010

2005

2030 2050

Europe

70

60

50

40

30

20

10

0

1950 1970 1990

Year

2010 2030 2050

Latin America and the Caribbean Northern America Oceania

70

60

50

40

30

20

10

70

60

50

40

30

20

10

70

60

50

40

30

20

10

0 0 0

1950 1970 1990 2010 2030 2050 1950 1970 1990 2010 2030 2050 1950 1970 1990 2010 2030 2050

Year Year Year

Source: World Population Prospects: The 2004 Revision, Datasets in Excel and PDF Formats, Extended Dataset, United

Nations, 2005.

N

OTE

: Estimates and medium-variant projections.

10

populations in 2050. In Oceania the equivalent increase is expected to be from 3 per cent to 7 per cent.

But it is in Asia and Latin America and the Caribbean where the increase is more remarkable: from barely

1 per cent in 2005 to 5 per cent in 2050. In contrast, the oldest old, which account for barely 0.4 per cent of Africa’s population today, are expected to account for just 1 per cent in 2050, indicating that Africa’s population will still be fairly young by mid-century.

Averages for major areas conceal large differences in the ageing process at the country level. Figure

4 shows the distribution of countries according to the percentages of children, adults of working age, and older persons, by major area in the developing world and for the more developed regions as a whole. The horizontal dark bars for each period indicate the central four-fifths of each distribution, the dot within the bars shows the median value, and the dots outside show the extreme values of each distribution (lowest and highest values observed). Focusing on the distributions relative to the proportion under age 15, a tendency for the bars and the highest value to move to the left is indicative of general reductions of the proportion of children and, therefore, of population ageing. Alternatively, a movement of the bars and lowest value to the right, indicates increases in the proportion of children and hence population rejuvenation. The diagrams thus make clear that many countries in Africa, Asia and Oceania, and Latin

America and the Caribbean first experienced some rejuvenation of the population before ageing set in.

For Asia and Oceania as well as for Latin America and the Caribbean the distributions begin to move to the left during the 1980s, but for Africa some leftward movement in the distribution occurs only after

2000. In all major areas, the initial movement of the distribution of the proportions of children to the left produces an elongation of the central bar, indicating that heterogeneity within the region increases as different countries embark on the transition to low fertility at different times. But, as time elapses, the length of the bar tends to diminish because more countries follow the path to ageing. Figure 4 shows clearly that there is greater homogeneity already among the more developed countries than in the developing regions. But both in Asia and Oceania and in Latin America and the Caribbean heterogeneity in terms of the proportion of children in the population is expected to decrease substantially by midcentury. The same is not expected for Africa, where even by mid-century there is likely to be considerable heterogeneity among countries in terms of their proportions of children.

The distributions of the proportion of persons aged 60 years or over contrast with those relative to the proportion of children because they tend to move in the opposite direction, as the proportions of older persons rise. In addition the diversity or variability among countries tends to increase as time elapses, so that, at least in the less developed regions, the dark bar is longer around 2050 than in earlier periods. It is noteworthy that the movement to the right of the distributions begins earlier in Latin America and the

Caribbean than in Asia and Oceania, and that it is hardly noticeable in Africa before 2010. For the more developed regions, the movement to the right of the full distribution is steady and was already under way by the late 1960s. This pattern together with the relatively narrow central bars suggests that the ageing process is pervasive and relatively homogeneous in the more developed countries.

Lastly, the distributions of the proportion of persons aged 15-59 illustrate the movement of the population wave that eventually leads to rapid ageing. In the less developed regions, the proportions 15-

59 tend to decline at first (moving the distribution to the left) as the population becomes younger when mortality decreases; then, as fertility declines, they increase and the distribution moves to the right. Later, at the third stage of the transition, the distribution moves again to the left as the proportions 15-59 decline in favour of the increasing proportions of older persons. In the less developed regions the distributions are still moving to the right as the proportions 15-59 generally increase. The movement to the left is not expected to occur before 2030 in Asia and Oceania nor in Latin American and the Caribbean.

Furthermore, in Africa the movement leftward is not likely to occur before 2050. In sharp contrast, the movement to the left is just about to start in the more developed regions and is expected to be persistent over the coming decades.

11

Figure 4. Distribution of the proportion of population aged under 15 years, 15-59 years and 60 years or over, 1950-2050, for major areas

Percentage under 15 years

More developed regions

a

Percentage aged 15-59 years

More developed regions

a

Percentage aged 60 years or over

More developed regions

a

1950

1970

1990

2010

2030

2050

2005

Less developed regions

Africa

Less developed regions

Africa

Less developed regions

Africa

1950

1970

1990

2010

2030

2050

2005

Asia and Oceania b

Asia and Oceania b

Asia and Oceania b

1950

1970

1990

2010

2030

2050

1950

1970

1990

2010

2030

2050

1950

1970

1990

2010

2030

2050

2005

1950

1970

1990

2010

2030

2050

Latin America and the Caribbean Latin America and the Caribbean Latin America and the Caribbean

1950

1970

1990

2010

2030

2050

2005

1950

1970

1990

2010

2030

2050

10 20 30

Per cent

40 50 40 50 60

Per cent

70 80

Source: World Population Prospects: The 2004 Revision, Datasets in Excel and PDF Formats, Extended Dataset,

United Nations, 2005.

Note: The figure summarizes the range and distribution of country values of percentages of the population aged under 15 years,

15-59 years and 60 years or over, at 10-year intervals from 1950 to 2050. Uniited Nations estimates and medium-variant projections.

a More-developed regions include Europe, Northern America, Japan and Australia and New Zealand.

b Excluding Japan, Australia and New Zealand.

0 10 20

Per cent

Minimum Median

30 40

Maximum

Central 80 per cent

12

Note that, especially in Asia and Oceania, there are some countries where the proportion aged 15-59 is projected to reach exceptionally high levels (above 70 per cent). The countries with these high proportions are mostly small and tend to be important destinations for migrant workers, such as the member States of the Gulf Cooperation Council. In addition to these countries, there are several others where the proportion aged 15-59 is already or is projected to be above 65 per cent. These include Algeria and Tunisia in Africa; Armenia, China, Iran, Mongolia, the Republic of Korea and Viet Nam in Asia; a number of the successor States of the former USSR including the Russian Federation, and several countries in Eastern and Southern Europe. All of these countries have experienced either sharp drops in fertility or long periods of very low fertility or both.

C.

POPULATION AGEING AT THE COUNTRY LEVEL

One useful indicator of population ageing is the median age of a population, that is, the age that separates the population into two equal halves. Increases in the median age are indicative of population ageing. Between 1950 and 2005, the median age of the world population increased from 24 to 28 years and it is projected to rise to 38 years by 2050 (table 4). More developed countries already have a median age of 39 years and it is projected to rise to over 45 years by mid-century. The less developed countries as a whole have a much younger population, with a median age that is currently 26 years and that is projected to be somewhat below 37 years in 2050. Particularly young populations are found in the group of 50 least developed countries whose overall median age is still under 19 years today and will likely not surpass 28 years over the next 50 years. These countries include most of those where fertility is still high and has shown no decrease or only an incipient decline.

T

ABLE

4. M

EDIAN AGE FOR THE WORLD

,

THE DEVELOPMENT GROUPS

AND THE MAJOR AREAS

, 1950, 2005

AND

2050

Regions and major areas

Median age (years)

1950 2005 2050

More developed regions

Less developed regions

Least developed countries

29.0 38.6 45.5

21.4 25.6 36.6

19.6 18.9 27.3

Africa

Asia

Latin America and the Caribbean

Northern America

Europe

19.0 18.9 27.4

22.0 27.7 39.9

20.2 25.9 39.9

29.8 36.3 41.5

29.7 39.0 47.1

Oceania 28.0 32.3 40.5

Source: World Population Prospects: The 2004 Revision,

Datasets in Excel and PDF Formats, Extended Dataset, United

Nations, 2005.

At the level of countries, the youngest populations are those of Mali, Niger and Uganda, whose median ages in 2005 were at most 16 years. Countries with young populations typically have still or had until recently high fertility levels. The map in figure 5 indicates that the youngest populations are mainly concentrated in the countries of sub-Saharan Africa. Young populations also characterize countries in the

Arabian Peninsula and a few countries in South-Central and Eastern Asia, including Afghanistan and

Pakistan. Most countries in Central America also have young populations as do Bolivia and Paraguay in

South America.

13

Figure 5. Median age by country, 2005

At the other end of t he spectrum, countries such a s German ages surpassing 42 years. Indeed, as figure 5 shows, all developed countries have m n already have median edian ages at or above

35 years. Among the less de veloped countries, those with the olde p o clude Argentina, Chile,

Cuba and Uruguay in Latin America as well as China, Kazakhstan and Thailand in Asia but most of them still have median ages b elow 35 years.

Over the next 45 y ears, most countries will see their median y. As the map in figure

6 shows, by 2050 very few countries are expected t o have median ages below 22 years. Burundi, Chad,

Congo, Guinea-Bissau, Guinea, Liberia, Niger and Uganda are projected to have the youngest populations at that time. In sharp c ontrast, 17 countries are expected to have her than 50 years, the majority located in Europe (e.g., Bulgaria, the Czech Republic, Italy, Lithuania, Poland, Ro mania,

Slovakia, Slovenia and Ukraine) or in Asia (Hong Kong SAR China, Georgia, Japa n, and the Republic of

Korea). As the map in figure 6 shows, by 2050 the median ages of all developed countries except the

United States are expected to be equal or greater than 42 years, and a number of developing countries are also projected to have median ages in that range, including China, Chile, Cuba, Kazakhstan, Mexico,

Thailand and Tunisia. Furthermore, with the exception of countries in sub-Saharan Africa, most of which continue to have relatively young populations in 2050, virtually all other developing countries are projected to have median ages at or above 35 years by mid-century. Among developed countries, the

United States maintains a relatively younger population because of international migration: without migration its median age in 2050 would be projected at 43.6 years instead of the 41.1 years projected with migration.

14

Figure 6. Median age by country according to the medium variant, 2050

D. I

S POPULATION AGEING INEVITABLE

?

So far, the discussion of future prospects has been based on the results of the medium variant of the official United Nations population projections (United Nations, 2005a). However, the projections include other projection variants aimed to illustrate the effects of different future paths of fertility change. In particular, the high and low variants differ from the medium variant in that their future fertility remains half a child above and half a child below that of the medium variant over the projection period (2005-

2050). These differences imply that, whereas in the medium variant the fertility of most countries reaches

1.85 children per woman by the end of the projection period in 2045-2050, in the high variant it reaches

2.35 children per woman and in the low variant 1.35 children per woman. That is, both the low and medium variants project eventual fertility levels that do not ensure the replacement of populations and would therefore lead to declining populations, whereas the high variant projects fertility levels that, if sustained, would produce a continually growing population. These assumptions lead to vastly different populations in 2050: the low variant produces a population of 7.7 billion persons that is already declining; the high variant produces a population of 10.6 billion that continues to increase and the medium variant produces a population of 9.1 billion that is still increasing but very slowly. It is expected that future population trends will be within the range projected, particularly because if fertility at the world level would differ markedly from the trends projected the population would growth faster or decline earlier and more rapidly than projected, outcomes that would more likely be unsustainable.

15

What effects do the fertility levels projected in the high and low variants have on the age distribution expected by mid-century? As table 5 indicates, at the world level the low variant projects a population where the proportion of children is 15 per cent and that of older persons is 26 per cent, while in the high variant the proportion under 15 is 25 per cent and the proportion aged 60 or over is 18 per cent. In comparison with today’s figures of 28 per cent under age 15 and 10 per cent aged 60 or over, the low variant produces a population that is ageing very rapidly. But even in the high variant, where fertility remains above replacement level and the population continues to increase, producing therefore only a small reduction in the proportion of children, the proportion of older persons nearly doubles, partly because of the expected continuation of mortality reductions, especially at older ages. Higher fertility as projected in the high variant is also insufficient to stop the ageing of the population of more developed regions, although it does increase somewhat the proportion under age 15 (it rises from 17 per cent in 2005 to 20 per cent in 2050). But the high variant still produces fairly marked ageing in the less developed regions. Thus, even under the assumption of continued population growth, population ageing seems inevitable.

T

ABLE

5. P

ERCENTAGE DISTRIBUTION OF THE ESTIMATED AND PROJECTED

POPULATION OF THE WORLD AND THE DEVELOPMENT GROUPS ACCORDING

TO DIFFERENT PROJECTION VARIANTS

, 2005

AND

2050

Age group

2005 2050

Estimates Low Medium High

0-14

15-59

60+

Total

Number (billions)

0-14

15-59

60+

Total

Number (billions)

0-14

15-59

60+

Total

Number (billions)

28

61

10

World

15

59

26

100 100

6.5 7.7

20 25

58 57

22 18

100 100

9.1 10.6

More developed regions

17 11 16 20

63 51

20 38

52

32

52

28

100 100

1.2 1.1

100 100

1.2 1.4

Less developed regions

31 16 21 25

61 60

8 24

100 100

59 57

20 17

100 100

5.3 6.6 7.8 9.2

Source: World Population Prospects: The 2004 Revision, Datasets in

Excel and PDF Formats, Extended Dataset, United Nations, 2005.

16

How about the effect of international migration? In today’s world, the more developed regions are net receivers of international migrants while the less developed regions are net senders. The medium variant assumes that the direction of those flows will be the same in the future. Therefore, the medium variant already reflects the possible rejuvenating effect that migration may have on the population of the more developed regions. To gauge the size of that effect, projections assuming that there is no migration have been calculated. They show that, without migration, the population of the more developed regions would be expected to have 15 per cent of the population under age 15, 50 per cent aged 15-59 and 35 per cent aged 60 or over. That is, a net migration gain averaging 2.2 million persons per year during 2010-

2050 would reduce the proportion aged 60 or over by 3 percentage points and increase the proportion of children by one percentage point, a fairly small contribution to the slowing down of the ageing process.

This finding is consistent with those of other studies that have examined international migration’s effect on population age structures and have concluded that the effect is usually small, though not always negligible (see United Nations, 1998; 2001). Although migrants tend to be concentrated in the young adult ages and are often accompanied by young children, thus adding to the younger age groups, in order for migration to have a lasting effect on the age structure of a population, a sustained inflow of migrants over time is required. Simulations to calculate the level of international migration necessary to offset the reduction of population numbers in countries with below-replacement fertility, suc h as Italy, Japan or the

Republic of Korea, have show n that the levels of immigration required w ould have to be much higher

1). To offset projected population decline in Europe, for instance, the net nu mbe r of immigrants would need to be twice as high annually during 2000 -2050 as it was during 199 5-2000. The levels of immigration needed to offset projected declines in the working-age population would be large r still.

Furthermore, even those levels of net immigration would not halt the process of populat ei Indee n migration flows would be required in countries with below-replacement fertility to kee p constant the ratio of persons over age 65 to the population of w orking-age. Such inflows w ls sult in y pulation growth driven mainly by the ne w imm igrants and their descendants. Such a scenario underscores the fact that the ongoing process of po pulation ageing has a powerful in o um t a be halted by major population increases which would in all likelihood be unsustainable.

E. T

H E

C

OMPONENTS OF CHANGE STRU CTU E

It is of interest to ascertain to what extent the changes in fertilit o y and international migration expected in the f uture have an impact on the eventua l ag u

Following a procedure used in United Nations (1988) w hereby differe nt p ti of the population. re made keeping all components of change constant, except one, so as to assess its effect, the estim ates present ed in table 6 were obtained. In general, the age distribution of a closed population is determined by the fertility and mortality to which the population has been subject. The initial ag e str uctur e of e po pulation also has an effect on the eventual a ge d istribution, but this effect d iminis he s as the projection period increases. In addition, if the population experiences non-zero migr th tter also affects the resulting age distribution.

As table 6 indica tes, in the more developed regions of children aged 0-14 declined by

9.1 percentage points betw een 1950 and 2000 and is projected to decline further by 2.7 percentage points from 2000 to 2050. D uring 1950-2000, fertility decline was responsible for m ost of the change observed.

In the future, however, because fertility is expected to recover somewhat from the very low levels it has reached in most developed countries, its contribution to the proportion of children in more developed regions is expected to be positive. However, such contribution is more than counterbalanced by the negative contributions of mortality and of the initial age distribution. The contribution of mortality is negative during 2000-2050 because mortality levels are expected to decline faster among older persons

17

T

ABLE

6. C

OMPONENTS OF CHANGE IN THE PROPORTIONS OF POPULATION UNDER AGE

15,

AGED

15-59

AND

60

OR OVER FOR THE DEVELOPMENT GROUPS

, 1950-2000

AND

2000-2050

Initial proportion (percentage)

More developed regions Less developed regions

1950200019502000-

2000 2050 2000 2050

27.4

0-14

18.3 37.6 32.9

Proportion at the end (percentage)

Absolute change

18.3

-9.1

15.6

-2.7

32.9

-4.7

20.9

-12

-5.8

0.2

-0.1

-6.3

Fertility effect

Mortality effect

Migration effect

Effect of the initial age distribution

Initial proportion (percentage)

Proportion at the end (percentage)

Absolute change

Fertility effect

Mortality effect

Migration effect

Effect of the initial age distribution

Initial proportion (percentage)

Proportion at the end (percentage)

Absolute change

-9.3

-0.1

0.3

0

60.9

62.3

1.4

5.2

-2.1

0.3

-2

11.7

19.5

7.8

2.1

-1

0.6

-4.4

15-59

62.3

52

-10.3

-0.4

-2.9

1.5

-8.5

60 or over

19.5

32.4

12.9

-11.9

2.8

0.1

4.3

55.9

59.4

3.5

9.9

-3.5

-0.2

-2.7

6.4

7.7

1.3

7.7

20

12.3

59.4

59.1

-0.3

3

-2.1

-0.1

-1.1

Fertility effect

Mortality effect

4.1

2.2

-1.7

4

2

0.7

2.9

2

Migration effect -0.4 -2.2 0.2 0

Effect of the initial age distribution 1.9 12.8 -1.6 7.4

In the more developed regions, fertility made a positive contribution to the growth the proportions

1 5-59 and 60 or over during 1950-2000, precisely because it had a negative effect on the growth of the prop ortion of children over the same period. Mortality made a negative contribution to the proportion 15-

59 while it contributed positively to that 60 or over because it declined faster at older ages. Similarly, the initial age distribution contributed to reduce the share of the population aged 15-59 but it raised that of the older population. By contrast, migration had a small positive effect on the share of those aged 15-59 but a negative one on the share of the older population. During 2000-2050, the 2000 age distribution in more developed regions has by far the major effect on the changes expected in the proportions aged 15-59 and

60 or over. In addition, mortality has an important role, raising the proportion 60 or over while reducing that 15-59, and migration has a moderate effect in the opposite direction in both instances. Fertility, which

18

has a positive effect on the proportion of children during 2000-2050, has consistently negative effects at ages above 15.

In the less developed regions changes in fertility are the dominant factor determining the change in the age distribution, followed by the effects of the initial age distribution, pa rticularly with respect to changes in the proportion under 15 during 1950-200 0 a nd in t he prop or ti on 60 or o ver d urin g 2000-2050.

The effect of mortality on the changing proportions u nder 15 an d 15 i gnifi cant durin g 1950-2000 and it is expected to make a moderate contribution to the proportion of older persons in the future. In the less develo ped regions, international migration had a v ery small c urin g 195 0-20 00 at its effect becomes vi rtually nil during 2000-2050.

In sum , fertility chan ge was an important contribut or to the g in the age distri butio n of both more developed and less developed regions in the past and is expected to continue having an important role in the less develope d regions in the future. In addition, the e s f the 2000 age distributio n on future deve lopment are substantia l in both groups of countries but especially in the m ore developed regions. Th at is, much of the expected future ageing of to day’s po pulation is already insc ribed in its age distribution which itself is the result of d istor ca ue transi tion from high mortality and fertility to low levels of both..

F. D

EPENDANTS AND WORKERS

Indicators of the potential effects of ageing for social and econom ic development are the so-called dependency ratios whic h relate the group of the popu lati on th at is co dependent to the group th at is economically active. Since exact measures of the number of pe rsons who are produc based solel ers and those who are exclusively consumers y on age ranges. Thus, it is g enerally assume

are n ailab le, pr oxies are used h n nder 15 f all in the dependent mainly consumers. In this paper, the older population has been defined as that aged 60 years or over. For consistency, that age group will be used here to calculate dependency ratios. Three types of ratios will be considered: (a) the child dependency ratio which relates the number of person s aged 0-14 to th e number aged 15-59 ; (b) the old-age dependency ratio which relates the n those aged 15-59, and (c) the overall dependency ratio which is the e f per

the ot sons her

aged two.

60

All d or over to ependency ratios are expressed in terms of number of children or older persons or both per 100 persons aged 15-59.

Before proceeding to an analysis of trends in these dependenc y r ti me c aveats are in order.

Clearly, in most populatio ns people do not stop being economically acti e 60, nor is it true that all persons age d 15-59 are economically act ive. Furthermore, althoug n en req uire economic s upport from others, older persons in many societies are also providers of support to their adult children

(M organ, Schuster and Butler, 1991; Saad, 2001). In addition, not all persons aged 15-59 provide direct or in direct support to children or older persons (Taeuber, 1992). In particular, as the period of training for a productive life increases, most adolescents and young adults remain in school and out of the labour force, effectively extending the period of young-age dependency well beyond age 15. These observations suggest that, at best, trends in the dependency ratios considered here are indicative of the constraints that society may face as the population ages and unprecedented imbalances arise in the sizes of key age groups. However, estimation of the likely economic impact of such imbalances would require more appropriate measures of the number of workers and consumers in a population by age.

Table 7 shows the values of the three dependency ratios for selected years from 1950 to 2050 for the world, the more developed regions and the less developed regions. As expected, the overall dependency ratio for the world increases from 74 dependents per 100 persons aged 15-59 in 1950 to 83 in 1975,

19

mainly as a result of the increase in the number of children relative to the working age population, as reflected in the rise of the child dependency ratio. Between 1975 and 2005, the overall dependency ratio d eclines from 83 to 63 dependents per 100 persons of working age, mainly because of the continued and majo r reductions in child dependency. However, this decline is expected to come to a halt in the coming decade so that, by 2025, the overall dependency ratio is projected to be 65, slightly higher than in 2005, and by 2050 it would reach 72. This projected increase in the overall dependency ratio is entirely due to the rise of old-age dependence since the ratio relative to children is expected to maintain its decreasing path (figure 7). It should be noted that these trends would be similar if the ratios considered used the population aged 15-64 as denominator and defined older persons as those aged 65 or over.

In the more developed regions, the overall dependency ratio increased slightly between 1950 and

1975 (from 64 to 66 dependants per 100 persons aged 15-59), and then declined to 59 in 2005, thus reaching a very low level and the minimum that it is expected to reach since a steadily increasing path is projected for the overall dependency ratio in the future. Such an increase results from the ever rising old- age dependency ratio. Indeed, by 2050 the old-age dependency ratio for the more developed regions is expected to reach 62, a value higher than today’s overall dependency ratio and close to the overall ratio estimated for 1950. Adding the dependent children produces an overall dependency ratio of 92 in 2050, a level 45 per cent higher than the average d ependency ratio prevailing in the more developed regions b etween 1950 and 2005. These trends confirm the earlier finding that the window of opportunity is essentially over for the more developed regions and that rapid ageing lies ahead.

T he experience of the less developed regions and their prospects are quite different. First, their over all dependency ratio in1950, at 79 dependents per 100 persons aged 15-59, was high compared to that of the more developed regions at the time and it was mostly due to the high level of child dependency (67 per 100). Between 1950 and 1975, as the proportion of children increased because of reduced mortality, both the child and the overall dependency ratios soared to 78 and 90 respectively. But the reduction in the proportions of children coupled with the rising proportions of persons aged 15-59 that occurred after

1975 led to a major reduction of the child and overall dependency ratios. By 2005, the overall dependency ratio in the less developed regions stood at 64 and is expected to keep on declining until it reaches 61 in

2015 before a slowly increasing trend sets in. Increases after 2015 are slow because the rapid rise expected in the old-age dependency ratio are counterbalanced by continued reductions in child dependency. By 2050 the less developed regions as a whole are expected to have an overall dependency ratio of 69, lower than it was in 1950, but with a very different composition than in earlier times since older persons are expected to account for about half of the dependency burden, up from 15 per cent in

T

ABLE

7. T

HREE TYPES OF DEPENDENCY RATIOS RELATING THE POPULATION OF CHILDREN

AND THE ELDERLY POPULATION TO THE POPULATION AGED

DEVELOPMENT GROUPS

15-59

FOR THE WORLD AND THE

, 1950-2050

Type of ratio 1950 1975 2005 2025 2050

World Overall

Children

Elderly

74

60

14

83

67

16

63

46

17

65

40

25

72

35

37

More developed regions Overall

Children

Elderly

64

45

19

66

40

26

59

27

32

76

28

48

92

30

62

Less developed regions Overall

Children

Elderly

79 90 64 63 69

67 78 50 42 35

12 12 13 21 34

20

Figure 7. Trends in the three types of dependency ratios for the world, the development groups and the major areas,

1950-2050

World More developed regions Less developed regions

100 100 100

2005

80

80 80

Overall ratio

60 60 60

Children

40 40 40

20

Old-age

0

1950 1970 1990 2010

Year

Africa

2030 2050

100

20

0

1950 1970 1990

Year

Asia

2010 2030 2050

100

20

0

1950 1970 1990 2010

Year

Europe

2030 2050

100

80

60

40

80

60

20

0

1950 1970 1990 2010 2030

Year

Latin America and the Caribbean

2050

100

80

60

40

80

60

20

0

1950 1970 1990 2010

Year

Northern America

2030 2050

100

80

60

40

80

60

20

0

1950 1970 1990 2010

Year

Oceania

2030 2050

100

40

20

40

20

40

20

0 0 0

1950 1970 1990 2010 2030 2050 1950 1970 1990 2010 2030 2050 1950 1970 1990 2010 2030 2050

Year Year Year

Source: World Po

D , E xtended Dataset, United Nations, 2005.

N

OTE

: Estimat es and me dium-variant projections.

Figure 7 also illustrates the paths followed over time by the vera dep den rat io and its child

America and the Caribbean both had very hig

’s major areas. Between 1950 and 1975, Africa and Latin y s, upper 90s but, whereas the overall dependency ratio for Latin America and the Caribbean had dropped markedly by

2005 (to 63), th i 8). With res pect to the future, the period of steady decline of the overall dependency ratio still lies ahead for Africa where it is expected to reach 63 by 2050.

21

Fo r Latin America and the Caribbean, in contrast, the expectation is for the ratio to stop declining around

2020 and then initiate a steady increase to reach a level of 7 3 by 2050. Furthermore, as figure 7 illustrates, after 2040 much of the increase of the dependency ratio will come from the rising weight of the older population.

Asia’s overall dependency ratio follows a path similar to that of Latin America and the Caribbean but at a lower level, so that it reaches a low value of 59 in 2005. The prospects for Asia are also mainly a steady though slow increase in the overall dependency ratio that sets in at around 2010. In Asia the oldage burden becomes higher than the child burden slightly earlier than in Latin America and the Caribbean but by 2050 both regions are expected to have similar overall dependency ratios (72 for Asia).

The other three regions, being part of the developed world, exhibit lower overall dependency ratios before 2005 than the major areas of the developing world. Europe in particular maintains very low overall dependency ratios during that period, especially because of its lower levels of child dependency. In addition Europe is the first region to exhibit a crossover of the child and old-age dependency ratios so that by 2005 the older population accounts for 57 per cent of its overall dependency burden which itself stands at 58 dependants per 100 persons aged 15-59. It is this early crossover that leads to a sharp rise in the overall dependency ratio of Europe, which largely parallels that of its old-age ratio. As a result, by 2050

Europe’s overall dependency ratio is expected to be a high 98, similar to that of Africa in the 1960s but determined mainly by a high proportions of older persons instead of high proportions of children.

For Northern America and Oceania, the overall dependency ratios follow similar trends, having peaked at around 1960 at a level of 78-79 and then declined to 59 and 63, respectively, by 2005. In both regions the dependency ratios are expected to start increasing within the next decade to reach by 2050 a level of 79 in Northern America and 75 in Oceania. It is noteworthy, however, that the share of older persons in the overall dependency ratio is expected to surpass that of children earlier in Northern America

(in 2015) than in Oceania (2030). In both major areas, as in Europe, the expected increase in the overall dependency ratio is mainly driven by the rising weight of the older population.

G. T

HE SIZE AND GROWTH OF MAJOR AGE GROUPS

So far the discussion has focused on the distribution of population by age group or the relative size of one group to another but one important consequence of the demographic transition is the veritable explosion of population growth and its eventual decline, processes that have affected differently the growth of the major age groups in the population. Overall, world population increased from 2.5 billion in

1950 to 6.5 billion in 2005, a gain of 4 billion, nearly 2.6 times as many people as were alive in 1950.

But, as table 8 indicates, about 1.6 billion of this increase occurred between 1950 and 1975 and 2.4 billion during 1975-2005. During the first period, 60 per cent of the population increase was concentrated in the a e g roups 0-14 and 15-24. In contrast, between 1975 and 2005, 56 per cent of the increase occurred i n t he form of growth of the population aged 25-59 and during 2005-2025 that age group will also account for the largest share of the population increase expected (54 per cent). After 2025, however, 66 per cent or two-thirds of the expected growth will be concentrated in ages 60 or over.

22

Age grou p

T

ABLE

8. A

BSOLUTE INCREASE AND AVERAGE ANNUAL GROWTH RATE BY AGE GROUP

,

THE WORLD AND DEVELOPMENT GROUPS

,

1950, 1975, 2005, 2025

AND

2050

1950-

1975

Increase (millions)

1975-

2005

2005-

2025

2025-

2050

Average annual growth rate (percentage)

1950-

1975

1975-

2005

2005-

2025

2025-

2050

Percentage distribution of the population increase by age group

1950-

1975

1975-

2005

2005-

2025

2025-

2 050

0-14

15-24

2559

6079

80+

Tot al

633

298

479

127

323

402

1 343

267

88

52

780

447

- 77

14

458

542

18 55 74 234

1 554 2 391 1 440 1 171

0-1 4

1524

2559

60-79

80+

Tot al

31

38

99

56

10

234

- 48

- 11

140

56

26

164

- 10

- 25

- 26

75

23

38

- 2

- 7

- 61

9

48

- 13

2.2

2.0

1.6

2.0

3.3

1.9

World

0.7

1.4

2.2

2.0

3.4

1.5

More developed regions

0.5 -0.7 -0.2

1.0

1.0

-0.2

0.9

-0.8

-0.2

2.0

3.1

1.0

1.1

2.9

0.5

1.6

2.1

0.2

0.2

0.2

1.2

2.8

3.1

1.0

-0.2

0.0

0.5

1.7

3.6

0.6

-0.1

-0.2

-0.4

0.1

2.1

0.0

41

19

31

8

1

100

13

16

42

24

4

100

14

17

56

11

2

100

-

-

-

-

-

-

54

31

6

4

5

100

-

-

-

-

-

-

0-1 4

1524

602

260

371

413

98

77

- 74

21

2559

60-79

80+

Total

380

70

8

1 203

211

29

806

371

50

518

532

186

1 320 2 227 1 403 1 183

Less developed regions

2.6 0.9 0.3

2.4

1.9

1.8

2.6

0.4

1.6

2.1

3.6

2.3

2.6

3.9

1.8

3.4

3.9

1.2

-0.2

0.1

0.6

2.1

4.4

0.7

46

20

29

5

1

100

17

19

54

9

1

100

57

26

4

100

7

6

Source: W orld Population Prospects: The 2004 Revision, Datasets in Excel and PDF Formats, Extended Dataset, United Nations, 2005.

Trends at the world level are determined mainly by those of the less developed regions so that the two resemble each other. Hence, population growth in the less developed regions during 1950-1975 was concentrated in ages 0-24, but during 1975-2025 its concentration shifted to age group 25-59. After 2025,

61 per cent of the expected population growth in the less developed regions is concentrated in age group

60 or over.

More developed regions display a different pattern of growth by age group. During 1950-1975, age group 25-59 absorbed the largest share of overall population growth, 45 per cent, whereas groups 0-24 and 60 or over accounted for almost equal shares of overall growth (28 per cent). During 1975-2005, however, the population under age 25 declined and population growth was concentrated in both age groups 25-59 and 60 or over, with the former accounting for the largest share. But in future even the population aged 25-29 is expected to decrease so that any population increase will be concentrated at ages

60 or over. That is expected to happen between 2005 and 2025 and, remarkably, also between 2025 and

2050 when overall population growth is projected to be negative even as the population aged 60 or over still increases substantially.

-6

2

44

45

16

100

-7

1

39

46

20

1 00

-

-

-

-

-

-

23

The trends described above result from the movement over the life cycle of the large cohorts born when fertility was high. In the developed world, they are related to the ageing of the “baby boom” generation, that is, the cohorts born fr

m en om the late 1940s to the 19 e arg or b

60s, when fertility rebounded. In the e coh ts born efore fertility declined significantly in t he late 1970s or early 1980s. The fact that fertility remains still high in an important p art of the developing world including most notably s ub-Sahara A pected increases in the s ize of ag e grou ps 0-24 an de op wor ld m ay s tart d 25-5

decli

9.

How ever, a ccord ing to c urr ent pr oje ction s, the child p ti th nin g afte 2 .

With regard to the speed of growth of different age gr oups, figure 8 displays the growth rates by fi ve-y ar a ge gr oup for the world and the development g rou ps. T he f igu o s ho w, for the worl d a nd

le de velop ed regio ns, r apid gro the youn ger ages gives way to rapid gro wth at the adult ages d t n to at der as ses.

It is noteworthy that pop o gro wth at t he o lde st a ges ov age 75 o r 80 ) rem ain s hig h th h ut the 1950-2 050 peri od. reduction of the population which first makes its appearance after 1975 when the younger age groups

(u r show ne gativ e growth. B y 2005-2 02 5 n ega tive g row th i s ex pect e d for all age ro el ag 5 by 2 ore developed regions, figure 8 m akes pla in t he p erva end ar e e vent ual

025 -205 0 al l grou ps b elow a ge 7 0 ar e ex pecte d t o be d ec reasin g i n si ze. y lion children and just under 1.2 billio n young persons aged 15-24, 88 per t who m l ive i n th e de velo ping wo rld. These numb ers ar e not expected to change much during the c omi ng decades (table 1). Bu t th eir d b ion amo ng the major regions will experience important shifts. us he oda y 60 per cent of the populati on u nder age 25 lives in , p A cent in Latin America and the Caribbean, and 12 per cent in the m ore developed regions, by 2050 the sh are f A frica is projected to ri se t p cen t, w hile thos e of Asi a, L A eri nd t he Carib an, d t mo re de vel oped regi ons will drop to 52, 8 an d 10 per cen t res pect ively.

Th e po pul ation aged 15-59 is proj ct o in crea se b y ab out one third etween 2005 and 20 50, p 2.

llion to 4 .1 b illi a e 1).

All ed ake e e l de vel ed ns e d e regi ons, the populati on of w n ag pr ed to s tart declining after 2005, so that by 2050 it will be 15 per cent smaller than in 2005 (510 million compared to

596 million). The decline will take place mainly in Europe, where the population aged 25-59 is expected to be almost 30 per cent lower in 2050 than in 2005, and in Japan, where the decline will amount to one third of the current number. In Africa, by contrast, the population aged 25-59 in 2050 is projected to be almo st two and a half times as large in 1950 as it is today: 838 million vs. 294 million (table 3).

The number of persons aged 60 or over is expected nearly to triple, passing from 672 million in 2005 to nearly 2 billion by 2050. The increase in the number of older persons is expected to be more marked in the less deve loped regions, where it may almost quadruple, passing from 428 million to nearly 1.6 billion during 2005-2050. The number of older persons will also rise in the more developed regions, but more mod erately, passing from 244 million in 2005 to 400 million in 2050. As a result, whereas today 64 per cent of the population aged 60 or over lives in developing countries, by 2050 the equivalent proportion is expected to be nearly 80 per cent. Today, 54 per cent of the older population lives in Asia and 22 per cent in Europe. The shares of Africa, Latin America and the Caribbean, and Northern America

24

Figure 8 Ave rage annual growth rate of population in 5-year age gr oups, for four time period s during 1950-2050: w orld and devel opme nt groups

A. World B. Mo re developed regions C. Less develop ed regions

1950-1975 1950197 5

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

0-4 5-9

10

-14

15

-19

20

-24

25

-29

30

-34

35

-39

40

-44

45

-49

50

-54

55

-59

60

-64

65

-69

70

-7

4

75

-79

Age

80

+

1950-1975

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

0-4 5-9

10

-14

15

-19

20

-

24

25

-2

9

30

-34

35

-39

40

-44

45

-49

50

-54

55

-59

60

-64

65

-69

70

-74

75

-79

Age

80

+

4.0

3.0

2.0

1.0

0.0

-1.0

0-4 5

-9

10

-14

15

-19

20

-24

25

-29

30

-34

35

-39

40

-44

45

-49

50

-54

55

-59

60

-64

65

-69

70

-74

75

-79 80

+

Ag e

1975-2005 1975-2005 1975-2 00 5

1.0

0.0

-1.0

5.0

4.0

3.0

2.0

2.0

1.0

0.0

-1.0

5.0

4.0

3.0

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

0-4 5-9

10-

14

1

5-19 20-

24

25-

29

30-

34

35

-39

40

-44

45-

49

50

-54

55

-59

60-

64

65-

69

70

-74

75

-79

Age

80

+

0-4 5-9

10-

14

15

-19

20

-2

4

25-

29

30-

34

35

-39

40

-44

45-

49

50

-54

55

-59

60-

64

65-

69

70

-74

75

-79

Age

80

+

0-4 5

-9

10-

14

15

-19

20

-24

25-

29

30-

34

35

-39

40

-44

45-

49

50

-54

55

-59

60-

64

65-

69

70

-74

75

-79 80

+

Ag e

2.0

1.0

0.0

-1.0

5.0

4.0

3.0

2005-2025

0-4 5-9

10

-14

1

5-1

9

20-

24

25

-29

30

-34

35-

39

40-

44

45

-49

50-

54

55-

59

60

-64

65

-69

70-

74

75-

79

Age

80+

2005-2025

2.0

1.0

0.0

-1.0

5.0

4.0

3.0

0-4 5-9

10

-14

15-

19

20-

2

4

25

-29

30

-34

35-

39

40-

44

45

-49

50-

54

55-

59

60

-64

65

-69

70-

74

75-

79

Age

80+

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

2005-2 025

0-4 5

-9

10

-14

15-

19

20-

24

25

-29

30

-34

35-

39

40-

44

45

-49

50-

54

55-

59

60

-64

65

-69

70-

74

75-

79 80+

Ag e

2025-2050 2025-2050 2025-2 050

2.0

1.0

0.0

-1.0

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

5.0

4.0

3.0

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

0-4 5-9

10

-14

1

5-1

9

20-

24

25

-29

30

-34

35-

39

40-

44

45

-49

50-

54

55-

59

60

-64

65

-69

70-

74

75-

79

Age

80+

0-4 5-9

10

-14

15-

19

20-

2

4

25

-29

30

-34

35

-39

40-

44

45

-49

50-

54

55-

59

60

-64

65

-69

70-

74

75-

79

Age

80+

0-4 5-9

10

-14

15-

19

20-

24

25

-29

30

-34

35-

39

40-

44

45

-49

50-

54

55-

59

60

-64

65

-69

70-

74

75-

79 80+

Ag e

Source: Wo

rld Population Prospects: Th e 200

4 Revision, Datasets in Ex cel an d PDF Formats, Exte nded D ataset, Uni ted Na tions, 2005.

25

are about 7 per cent each. By 2050, Asia is expected to have the largest share of the older population (63 per cent), whereas Africa, Europe and Latin America and the Caribbean will have about 10 per cent each and Northern America’s share would decline to 6 per cent.

A notable aspect of the global ageing process is the progressive demographic ageing of the older pop ul atio n itself. In most countries, regardless of their geographic location or developmental stage, the pop ul atio n aged 80 or over is growing faster than any younger segment of the older population. At the glo ba l lev el, the average annual growth rate of persons aged 80 years or over in 2000-2005 (4.2 per cent) was more than twice as high as the growth rate of the total population aged 60 or over (2.0 per cent).

Alt ho ugh those growth rates are expected to decline over the next 45 years, by 2045-2050 the growth rate of p op ula tion aged 80 or over (3.0 per cent) will still be almost double that of the group aged 60 or over

(1.7

p er c ent). Such high rates of growth produce in 2050 almost 400 million persons aged 80 or over, nea rly fiv e times as many as in 2005 (87 million). Currently, a slight majority of the oldest old live in the mo re developed regions. By 2050, over 70 per cent are projected to live in the less developed regions.

Although the proportion of the oldest-old is growing, they will remain a minority among older per sons. Today, just 12 per cent of the older population is aged 80 or over and in 2050 the projected num bers show that about 20 per cent could be in that age group. Worldwide, in 2050, nearly half (47 per cen t) of the older population will be aged 60-69 and one third will be aged 70-79.

H. T

HE DISTRIBUTION BY SEX OF THE OLDER AGE GROUPS

N orm ally, more boys are born than girls and the average sex ratio at birth is a biological constant that var ies lit tle among populations (it usually ranges from 103 to 106 boys for every 100 girls). Hence, diff er ence s in the sex ratio by age are determined by differential mortality by sex. Because wo men usu all y h ave a survival advantage over men, they tend to outnumber men as age advances. Thus, whe reas it is e stim ated that there are today 101 males per 100 females in the world population, the equivalent ratio is 8 2 men per 100 wom en among those aged 60 or over (table 9). This means that women aged 60 or over out nu mbe r men in the same ages by about 67 million. Furthermore, the proportion of women in the older pop ulation grows substantially with advancing age. In 2005, women outnumber men by almost 4 to 3 a t age s 65 or older, and by almost 2 to 1 at ages 80 or over.

Mortality levels have dropped markedly since 1950. Between 1950-1955 and 2000-2005 the world pop ulation gained 18 years of life expectancy, but the gains were not equally shared by men and women

(tab le 10). In most countries, reductions in mortality have been greater for females than males, thus rein forcing the female survivorship advantage. At the world level the female advantage in life expectancy at b irth has increased from 2.8 to 4.5 years since 1950. Furthermore, in terms of life expectancy at ages 60 and 80, the female advantage has also risen. Consequently, not only are women more likely than men to sur vive to age 60 but, once they reach that age, they can expect to live considerably longer than me n of the sa or o ve me r (8 age. These trends have contributed to maintain the low sex ratio of the world population aged 60

2 men per 100 women). Furthermore, among those aged 80 or over, the sex ratio has decreased ma rke dly , passing from 61 in 1950 to 55 in 2005.

Sex ratios at older ages tend to be lower in the more developed regions than in the less developed reg ions, partly because the sex advantage in female survivorship has been larger in the more developed reg ions, particularly after age 80. As a result, there are today 72 men per 100 women aged 60 or over in the more developed regions and just 46 men per 100 women aged 80 or over. Furthermore, given that the fem ale advantage in life expectancy in the more developed regions is expected to decline by 2045-2050, th e se x ratio of their population aged 60 or over is projected to increase between 2005 and 2050, passing from 72 to 78 males per 100 females. Increases are also expected in the sex ratio of those aged 80 or over.

26

T

ABLE

9. S

EX RATIO OF THE POPULATION IN SELECTED AGE

GROUPS

,

WORLD AND DEVELOPMENT REGIONS

, 1950, 2005

AND

2050

Age group

Sex ratio

(males per 100 females)

1950 2050

0-14

15-59

60+

80+

Total

104

100

80

61

100

2005

World

105

103

82

55

101

105

104

85

61

99

0-14

15-59

60+

80+

Total

More developed regions

103 105 105

90

74

58

91

100

72

46

94

103

78

57

94

0-14

15-59

60+

80+

Total

Less developed regions

105 105 105

106

86

103

88

104

86

68

104

66

103

63

100

Source: World Population Prospects: The 2004 Revision,

Datasets in Excel and PDF Formats, Extended Dataset, United

Nations, 2005.

In contrast, in the less developed regions, sex ratios in both age groups are expected to decline, indicating that there will be an exacerbation of the imbalance between men and women at older ages in the developing world.

Among the world’s major areas, Europe has today the lowest sex ratios at older ages: 69 men per

100 women aged 60 or over, and 42 men per 100 women aged 80 or over. These low ratios result from both large sex differentials in life expectancy and the long-term effects of the massive loss of young men during the Second World War.

27

T

A BLE

1 0. Life e xpectancy at birth and at ages 60 and 80 by sex, world and development groups, selected periods

1950-1955

Bo th sex es Femal e Male Difference

2000-2005

Both sexes Female Male Difference

Both sexes Female

2045-2050

Male

World

More developed regions

Less developed regions

46.6

66.1

41.1

48.

0

68.

5

42.

0

45.3

63.5

40.3

2.8

5.0

1.7

Life expectancy at birth (years)

65.4 67.7 63.2 4.5

75.6

63.4

79.3

65.2

71.9

61.7

7 .4

3.5

75.1

82.1

74.0

77.5

85.0

76.2

72.8

79.1

71.8

Difference

4.7

5.9

4.

4

World

More developed re gions

Less developed reg ions

14.8

18.0

13.0

15.

5

19.

6

13.

3

13 .9

16 .2

12 .6

1.5

3.4

0.6

Life expectancy at age 60 (years)

19.2 20.7 17.5 3.2

21.0

18.1

23.0

19.3

18.7

16.8

4.3

2.5

22.4

25.0

21.8

24.3

27.3

23.5

Life expectancy at age 80 (years)

World

More developed re gions

5.7

6.7

5.

7.

7

2

5

6

.4

.1

0.3

1.1

7.5

8.3

8.1

8.9

6.7

7.2

1.4

1 .7

9.0

10.4

Less developed reg ions 4.9 4.

8

Source: World Popul

ation Prosp ects: T he 2004

4 .9 0.0 6.8 7.2 6.3 0.9 8.5

Revision

, Data set s in Excel and PDF F ormats, Extend ed Dat set, Un ited Nation s, 2005.

9.9

1 1.5

9.3

20.5

22.6

20.0

7.

9

9.0

7.5

3.

8

4.

7

3.

5

2.

0

2.

6

1.

7

28

I. T

H E CH ANGING AGE STRUCTUR ES O F URBAN AND RURAL POPULATIONS

As populatio ns a round the world are growing older, they are also becoming more urban. The age structures of urban and rural populatio ns usu ally differ appreciably, mainly because urban populations have generally un der gone the tran sition to lo w m ortality and low fertility earlier than rural populations.

Therefore, urban populations are usually further along the path to population ageing than rural populati ons. However, urban areas are al con centrated in th e yo ung er ages a nd w ould so

th ma jor a ere fore ttractors of rural-urban migrants, who tend to be hel p in m oderating population ageing.

Using data gathered by population censuses supplemented in a few cases by information collected by nat ionally repre sentat ive surveys i t was possible t o ob g stribution of the population classified by u rban or rural area of residence for at least two points in tim e fo r 119 of the 192 countries or areas of the ad 200 firs t data set refer s to a d

0 a popu ate w ithi latio n the n of 100,0 per iod 19

00 o

60-1 r more. F

980 and t h or a lmost all the countries with data, the e se cond to the period 1981-2005. Those per iods were sele cted so as to ma ke po ssib le an a ssessment of changes in the age distributi ons of urban and rural populations from about 19 70 t o the 1990 s. Th e countries with data available have a combined pop ulation of 5.2 billio n in 2000, tha t is, they acco unt fo r 85 per cent of the world population at the time.

Tab le 11 presents the d ist rib

No te tha t co verag e is s om ut ion of countries with da ta availab le b y development group and major area. ew ha t be tte r fo r develo pe d th an for de velo ping countries (68 per cent vs. 60 per cen t of c oun tries have dat a, respecti vely ) and that , in terms of major area, co verage ranges from 50 per cen t of c oun tries i n O c ean ia to 68 per ce nt of thos e in E urope an d No rthern America.

To obtain the proportions of the population in each age group by rural and urban residence, the data ava ilable for co untrie s w ithin eac h ma jor area w ere added up wit hout adjustment for reference dates, com plet eness of c ove rage or variations in the definition of urban . In reporting the aggregates, the median ref erence date of the data in each aggregate is reported as an indicator of the reference date of the sum.

De s e encie s, th e estimate s o btained pr ovid e an i ndication of ho w age structures vary in urb an and rural areas.

According to United Nations estimates (United Nations, 2004), the world population will become hal f urban before 2010. However, the proportion urban varies m ark edly among major areas. Despite the tren d to war ds rap id urb anization , the populatio ns of Africa and Asia remain mainly rural and the pro porti on ur ban in those areas is still a low 40 per cent in 2005. Latin America and the Caribbean, in con trast, is highly urbanized with close to 78 per cent of its population living in cities or towns. Europe,

No rthern America and Oc eania are also highly urban, with Euro p e an d Oceania being 73 per cent urban in

200 exa

5 an A in u rban areas. It is in this context that an mination of differences in the age structures of urb an and rur a l populatio ns is to be examined. san

Given that urban popu lat ions the wo itation, vaccines and new drugs that red rld uce ove

m r have bee ortality and n t h e fi rst to benefit from improvements in

th a t urb an dwellers have been more adept at controlling their fertility, th e t ran sition to low mortality an d l o w fertility usually starts much earlier in urban than in rural areas. Consequently, it would be ex pec ted that the urban population would be older than that in rural areas. Ho we ver , as mentioned above, the age distribution of urban populations is also influenced by migration and because the net gains from rural-urban migration tend to be high at young adult ages, it is expected that urban populations may be som ew h at y ounger than they would have been in the absence of migration.

29

T

ABLE

11. A

VAILABILITY OF DATA ON THE POPULATION BY AGE GROUP AND URBAN

/

RURAL

RESIDENCE

Major area

World

More developed regions

Less developed regions

Number of countries

Total a

With data

192

44

148

119

30

89

Percentage with data

62

68

60

Africa

Asia

Latin America and the Caribbean

Europe and Northern America

Oceania

54

50

35

41

12

31

31

23

28

6

57

62

66

68

50 a: Only countries or areas with more than 150,000 inhabitants in 2000 are included.

Table 12 presents the age distribution of the urban and rural populations of the world and major areas by broad age groups. Two sets of data are presented in each case: one referring to the early 1970s and the second referring to the1990s or later. Note that, particularly for the most recent estimate, median reference dates vary from 1990 to 2001. In particular, considerably more recent data are available for the more developed regions than for the less developed regions.

Focusing first on change over time, it is clear that the proportion of children has decreased almost universally, whether in urban or rural areas The sole exception is among the rural population of Africa and this finding is consistent with the delayed transition to low fertility in that major area. It is noteworthy, however, that a small reduction in the proportion of children is noticeable for the urban po pulation of Africa, consistent with an incipient transition to lower fertility.

The proportion of the population aged 15-24 has also decreased in most areas over time. Exceptions arise again in the case of Africa but also for Latin America and the Caribbean where the proportion in age group 15-24 in urban areas remains unchanged and that in rural areas increases slightly. The effect of net rural-urban migration may contribute to the lack of change in the proportion for urban areas and because the data for Latin America and the Caribbean refer to earlier reference dates than those for Asia, the trends observed are consistent with a transition to low fertility that started in the 1970s and has not yet had an effect at ages above 15.

Regarding trends in the proportions aged 25-59 and 60 or over, increases are the rule except in

Africa where the estimates show no change in the proportion aged 60 or over whether in rural or in urban areas and an increase in the proportion aged 25-59 in rural areas, indicative of the ongoing rejuvenation of the population as fertility remains high.

Overall therefore, the trend toward population ageing is evident in both the urban and the rural populations of all major areas except Africa. From around 1970 to the most recent reference date, the age distributions of the urban and rural pop ulations of all areas except Africa shifted from the two younger age groups, especially from ages 0-14, to older ages, primarily as a result of declining fertility.

30

T

ABLE

12. P

ERCENTAGE DISTRIBUTION OF THE POPULATION IN URBAN AND RURAL AREAS BY MAJOR AREA

,

FOR TWO REFERENCE DATES

Major area Year

Percentage of the urban population in each age grou p

0-14 15-24 25-59 60+

P erc entage of the rural population in each age group

0-1 4 15-24 25-59 60+

World

More developed reg ions

Less developed re gions

Africa

Asia

Latin America and the Caribbean

Northern America

Europe

Oceania

1971

1996

1970

2001

1972

1994

1974

1991

1971

2000

1971

1993

1970

2000

1970

2001

31

25

24

19

35

28

40

38

32

25

39

30

28

21

22

18

20

18

18

15

21

20

21

21

21

19

20

20

18

14

18

15

41

46

44

49

38

44

34

36

41

47

35

42

40

49

46

49

9

1 1

1 3

1 8

9

7

9

14

16

6

8

4

5

7

14

17

13

39

34

27

20

41

35

45

46

40

33

46

38

31

21

26

21

17

17

15

14

18

17

18

19

18

17

18

19

16

12

15

14

36

40

41

45

35

39

31

30

36

41

30

35

39

49

42

43

8

9

16

21

7

8

6

6

7

9

6

8

14

18

17

21

1972 29 17 41 36 16 38 9

1990 23 17 45 15 29 15 44 11

Turning now to the level of urbanization of the population in each age group, the estimates derived from the countries with data available suggest that, at the world level, children 0-14 are less likely to be urban dwellers than persons in other age groups (table 13). This finding is consistent with the conclusions of earlier studies (United Nations, 1980; Hermanova, 1999). The proportions urban for other age groups are remarkably similar indicating that the population aged 25 or over is already 49 per cent urban.

In the more developed regions, the highest levels of urbanization are found among age groups 15-24 and 24-59, which are 75 per cent urban. Children are 72 per cent urban and older persons just 70 per cent.

These estimates reflect mostly the pattern in Europe, where children and older persons also appear to be less likely than persons in other age groups to live in urban areas. However, the data for Europe may be biased toward the more rural countries in the continent since the overall level of urbanization for Europe shown table 13 is somewhat lower than the 73 per cent estimated for the whole population of Europe in

2000 (United Nations, 2004).

The estimates for the less developed regions also show that children and older persons are less likely to be urban dwellers than persons aged 15-59. This pattern results mainly from those prevalen t in Africa

31

and Asia because in Latin American and the Caribbean there is not a large difference between the proportions living in urban areas for ages 15-59 and 60 or over.

T

ABLE

13. P

ERCENTAGE OF

Majo r area

I IN URBAN AREAS

,

THE W OR LD AND M AJOR A REA S

Ye ar

Pe rcentag e in urb an area s

0-14 15-24 25-59 60+ To tal

More develope d regions 1970

20

1972

63

7

23

70

75

29

67

27

62

70

24

66

74

26

25

27

20

1971 5 4 61 62 63 59

Northern Americ a 1970 72 76 74 73 74

62

1990 79 84 83 87 83

In sum, aside from showing that the degree of urbanization varies by age group, the estimates available also show that the large majority of the child population in Africa and Asia lives in rural areas, as does the majority of the older population of those major areas (about 70 per cent in Africa and 60 per cent in Asia). It also appears that about a third of the child and the older population of Europe still lives in ru ral areas, though the countries with data available may be less urbanized than those excluded. Lastly, in Latin America and the Caribbean, Northern America and Oceania, very high percentages of the whole population but also of children and older persons are already urban dwellers. These differences by major areas have important implications for the strategies that society may adopt to adjust to the expected changes in age distribution.

J. L

IVING ARRANGEMENTS OF OLDER PERSONS

As this paper has documented, populations everywhere are growing older. Furthermore, not only are more people surviving to reach old age, but those who attain old age are living longer than ever before.

32

As a consequence, families comprising three or even four generations have become common, considerably expanding the alternatives for living arrangements of older persons. At the same time, the social and economic transformations that many countries are undergoing are modifying many aspects of day-to-day family life, including traditions favouring lifelong co-residence of parents and children as a basic means of e urin g sup port for young and old.

A recent study by the United Nations presents n analys is of the patterns in the li ving arrangements of older persons based on comparable data for more than 130 countries (United Na tions, 2005b). The study finds that about one out of every seve n older persons or approximately 9 e live alone. The large majority of these people, some 60 million, are wome or ide , 19 per cent of women aged 60 or over live alone, compared with 8 per cent of men in the sa me age gro up (tab le 14). Older wo men are more children or grandchildren, whereas old er men are e ntly ma rried. Thus, about 45 per cent of women aged 60 y ears or over have a li ving spouse, wher eas nea rly 80 per cent of m en in that age group do. Older women are also more likely to live with grand -ch ildr en o nly or with relatives other than their k ly l e t a

T

ABLE

14. P

ROPORTION OF PE

WHO LIVE ALONE

:

RSONS AGE D

ESTIMATES FOR THE W

60

ORL

YEA

D A

RS O R OV

ND M AJOR

ER

AREAS

,

BY SE X

(Pe rcentage)

Major area or region Total Men

World ...................................

......

Africa ...................................

......

Asia ......................................

......

7 5 9

Europe........................................

26 13 35

Latin America and the

Caribbean ...........................

......

9 7 10

Northern America ......................

26 15

Oceania a

...............................

......

25

34

of Older Persons Around the

publication, ST/ESA/SER.A/

World.

(Un ited Nat ions

240).

a

Imputed on the basis of data that covered under half of the region’s population.

In the more developed regions, about a quarter of older persons live alone, but in the less developed regions just under 10 per cent do so. The proportions living alone are lowest in South-Eastern and South-

Central Asia, where about 5 per cent of older persons live alone, and highest in Northern and Western

Europe, where about a third of older persons do so.

Whereas the likelihood of solitary living at older ages had been increasing in many developed countries, in recent years that trend has slowed down or halted in some countries, including Canada, Italy and the United States. The reasons for such slow down involve a combination of longer life, which tends to decrease the prevalence of widowhood, declines in the proportion who ne ver married, and an upward trend in the age at which children leave home.

33

Older persons the world over are opting for independent living by either living alone or only with a spouse. This preference is pervasive in economically developed countries and it is growing in some developing countries. However, whereas the most common arrangement in developed countries is for older persons to live apart from their children, around 75 per cent of older persons in developing countries live with a child or grandchild (figure 9). In European countries, by contrast, just 25 per cent of older p ersons live with their children.

Figure 9. Living arrangements of older persons in more developed and less developed regions and selected major areas

(Persons aged 60 years or over)

More developed regions Less developed regions

With spouse only (13%)

Alone (25%)

Alone (7%)

With spouse only (43%)

With children/ grandchildren

(75%)

With children/ grandchildren

(27%)

Other relatives or nonrelatives (5%)

Other relatives or nonrelatives (5%)

Africa Asia

With spouse only (9%)

Alone (8%)

With spouse only (16%)

With children/ grandchildren

(74%)

Other relatives or nonrelatives (8%)

With children/ grandchildren

(74%)

Alone (7%)

Other relatives or nonrelatives (4%)

Latin America and the Caribbean

Europe

With spouse only (16%)

Alone (26%)

With children/ grandchildren

(62%)

Alone (9%)

With spouse only (43%)

Other relatives or nonrelatives (14%)

With children/ grandchildren

(26%)

Source: nited Nations (2005). Living Arrangements of Older

Persons Around the World

(United Nations publication,

ST/ESA/SER.A/240).

N

OTE

: Based on the population in households.

Other relatives or nonrelatives

(4%)

34

At the world level, the proportion of older persons who live with a child has been declining. This is so, even as the percentage of older persons who have living children has been increasing. In developed countries, older persons typically experience an “empty nest” phase of life, so that the percentage of older persons living with children declines substantially as age advances from the sixties to the seventies and the youngest children mature and leave home. By contrast, in many developing countries the percentage of older persons living with a child remains high at older ages, suggesting a predominant pattern of lifelong co-residence with at least one child.

Countries with higher levels of soc ial and economic developm ent have lower levels of co-residence with children. National levels of development explain much of the variation in living arrangements of older persons around the world. This suggests that, as levels of income and education rise, and as fertility decline leads to a smaller number of offspring, living arrangements in the developing countries might come to resemble those of the more developed regions. However, cultural factors may have long-enduring effects on living arrangements. For example, even after taking development levels into account, older

Europeans are less likely to live with children than are older persons in the less developed regions.

Within developing countries, higher socio-economic status is often associated with higher, not lower, levels of co-residence with children. More specifically, in countries with low levels of development, coresidence with children tends to be associated with higher social and economic status. Among countries at moderate levels of development, these differentials tend to disappear or even reverse direction, and in economically advanced countries, older persons living as a separate couple tend to have a higher socioeconomic status than those living with children.

Co-residence with children is an important element of the flow of support between family members.

This is particularly so with respect to informal support that depends on physical proximity, such as providing assistance with basic activities of daily living. For older persons living with a spouse, support in performing activities of daily living is primarily provided by the spouse. Furthermore, it is wives rather than husbands who usually provide such support.

In many developed countries, institutional living has become an option for older persons who have difficulty managing on their own or who need specialized medical services. The issue of how to provide long-term care for older persons who need assistance and the escalating costs of providing such care have become pressing policy concerns in developed countries. In some countries, policies promoting community support so that it is possible to “age in place” appear to have halted or reversed earlier trends towards higher rates of institutionalization. The main factor accounting for differences in levels of institutionalization among countries appears to be a society’s ability to support the costs of institutional care. Recent trends in a number of developed countries seem to indicate that institutional care is increasingly reserved for those with the greatest need for more intensive care, who tend to be the oldest and the frailest. In most countries, the level of institutionalization is higher for women than for men.

Women’s greater likelihood of being widowed is the main reason for their greater likelihood of institutionalization. Indeed, unmarried older men are usually more likely to live in an institution than unmarried women.

In countries with high rates of HIV infection, the proportion of older persons living with grandchildren, but not with children (skipped-generation households), has increased. In the countries where at least 10 per cent of adults have been infected with HIV, the proportion of older persons in

O lder women are more likely than older men to live in these skipped-generation households.

35

In developing countries, older persons living either alone or in skipped-generation households tend to be an especially disadvantaged group and older women are more likely than older men to live in those situations. Over 10 per cent of older women live in skipped-generation households in most countries of sub-Saharan Africa and in some of Latin America and the Caribbean. In Malawi, Rwanda, Zambia and

Zimbabwe, between 21 per cent and 34 per cent of older persons live in skipped-generation households.

Overall, there is great variability in living arrangements involving older persons. For instance, the proportion of older persons living with a child or grandchild ranges from 4 per cent in Denmark to almost

90 per cent in Bangladesh, while the proportion living alone varies from less than 1 per cent in Bahrain to alm ost 40 per cent in Denmark. Living arrangements are both determined by and determine the wellbeing of older persons. As populations continue to age, societies will have to develop novel ways of improving and promoting the self-reliance of older adults, perhaps by providing innovative services to permit older persons to remain living independently for as long as possible and by encouraging family cohesiveness and the continued involvement of family members in the care and well-being of the older generations.

K. C

ONCLUSION

In 2005, most countries of the world are far advanced in the demographic transition. Developed countries have already reached the threshold of the third stage of the transition, when population ageing is pervasive and rapid. The majority of developing countries are in the second stage, when a favourable age distribution gives rise to a demographic bonus. There remain, however, some countries that are only at the start of the transition to low fertility and fewer still where fertility decline has not yet started. Those c ountries, most of which are located in Africa, are expected to continue having a youthful population for som e time and may not enter the period of the demographic bonus until later in this century.

The changes set in motion by the demographic transition lead to an ongoing transformation of populations which entails both opportunities and challenges for development. In order to benefit from the demographic bonus countries need to adopt sound m acroeconomic policies to foster investment and job creation. Already the cases of the “East Asian Tigers” prove that the demographic bonus can contribute to a ccelerate economic growth (Bloom and Williamson, 1998). The issue is whether other countries undergoing the transition can reap similar benefits.

In developing countries, the demographic bonus appears after a period of rapid population growth where the population becomes younger. Consequently, both the number and the proportion of children and young adults rises. Countries that are just entering the second stage of the transition also have to face the need to educate the large number of children and young adults in their population. At the world level, the number of persons aged 15-24 is at an all time high of nearly 1.2 billion and this number is not expected to decline before 2050. Nearly 9 out of every 10 persons aged 15-24 live in the developing world. Providing them and the 1.8 million children of the world with adequate education and health services to enhance their productive role in society remains a major challenge.

Equally import ant is the need to start planning for the time when the demographic bonus ends and population ageing accelerates. Already today, when developing countries have a relatively youthful pop ulation, 64 per cent of the population aged 60 years or over lives in developing countries. In 20 years time, developing countries will be the home of 71 per cent of the world’s older persons. Although older persons will constitute just about 13 per cent of the population of the less developed regions in 2025, their numbers will soar to 850 million. Ensuring that these growing numbers of older persons have adequate support in old age, access to decent employment for those who need or want to remain e conomically active, and appropriate health care is also likely to prove challenging.

36

In the more developed regions, the rapid population ageing expected demands societal and economic adaptations to a new reality. Older people in developed countries remain in good health for longer than ever before and therefore have the potential to continue being productive. Increases in retirement age are already being instituted and will have to continue if mortality across the developed world keeps on falling as expected.

Future changes in age structure can be anticipated with considerable accuracy, partly because the older persons of mid-century have already been born. Policy makers and planners have therefore ample warning of what lies ahead. Although population ageing is inevitable, its consequences will depend on the measures developed to address the challenges it poses.

37

R

EFERENCES

Bloom, D.E. and J.G. Willliamson (1998). Demographic transitions and economic miracles in emerging Asia, World Bank

Economic Review, vol. 12, No. 3, pp. 419-456.

Herm anova, Hana (1999). Ageing populations in rural areas. pp. 192-197 in Ageing and Health: a Global Challenge for the

Twenty-First Century. Proceedings of a WHO Symposium (Kobe, Japan, Centre f. H. D. [compiled by] World Health

Organization [WHO]. Kobe, Japan: World Health Organization.

Morgan, D. L., T. L. Schuster and E. W. Butler (1991). Role reversals in the exchange of social support. Journal of

Gerontology, vol. 46, No. 5.

Saad, P. M. (2001). Support transfers between elderly parents and adult children in two Brazilian settings (CD-ROM). Paper presented at the Twenty-fourth IUSSP General Population Conference, Salvador de Bahia, Brazil, 18-24 August 2001.

Paris: International Union for the Scientific Study of Population.

Taeuber, C. M. (1992). Sixty-Five Plus in America. Current Population Reports, Special Studies P23-178. Washington, D. C.:

United States Bureau of the Census.

United Nations (1980). Patterns of Urban and Rural Population Growth, Sales No. E.79.XIII.9.

United Nations (1988). Economic and Social Implications of Population Aging. Proceedings of the International Symposium on

Population Structure and Development, Tokyo, 10-12 September 1987. Sales No. E.90.XIII.18.

United Nations (1998). World Population Monitoring 1997: International Migration and Development. (United Nations publication, Sales No. E.98.XIII.4)/

United Nations (2001). Replacement Migration: Id it a Solution to Declining and Ageing Populations? (United Nations publication, Sales No. E.01.XIII.19).

United Nations (2004). World Urbanization Prospects: The 2003 Revision. (United Nations publications, Sales No. E.04.XIII.6).

United Nations (2005a). World Population Prospects: The 2004 Revision, vol. I: Comprehensive Tables (United Nations publication, ST/ESA/SER.A/244); vol. II: Sex and Age Distribution of Populations (United Nations publication,

ST/ESA/SER.A/245).

United Nations (2005b). Living Arrangements of Older Persons around the World (United Nations publications,

ST/ESA/SER.A/240).

38

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement